GTA
All Springer/NP/PCP Air Gun Discussion General => PCP/CO2/HPA Air Gun Gates "The Darkside" => Support Equipment For PCP/HPA/CO2 => Topic started by: rsterne on April 07, 2020, 04:13:00 PM
-
I have been corresponding with Ron Burnett, who is working with his LabRadar on trying to develop Drag Coefficient curves for pellets.... Ron has tested both .177 cal and .22 cal pellets, and after some distance downrange he is finding that the data becomes inconsistent.... He shoots over an open field with no obstructions, and a long distance behind the target before any trees.... He has determined that when the Signal-to-Noise Ratio (SNR) drops below about 20 is when the data becomes inconsistent.... In his setting, that is at about 50-55 yards for .177 pellets and at about 75 yards for .22 cal.... even though there is a radar return at much longer ranges....
I would suggest that you keep this in mind if you have a LabRadar
. The range at which this occurs will depend on your surroundings, and the presence of anything that might disrupt or reflect the radar.... The SNR does not drop smoothly, but you will see a general decrease with range.... You may get the occasional SNR below 20 while most are still above.... That data may, or may not, be good data.... but once most of the SNR levels (which are reported in the detailed shot string downloads) are below 20, your data may be questionable.... This most frequently shows up as velocity readings that are faster than the preceeding one, which of course is impossible.... and leads to negative Cd calculations (again, impossible) for single data pairs
.
Bob
-
Thanks for that update, Bob.
You say things the way they are. 👍🏼I've recently gotten a little tired of reviewers where anything they test is "badas$$" or where they mostly "test" products from one brand, and similar games....
So, no .22 pellet BCs beyond 75y... disappointing, as it may be to me and many others.... 😟
Well, I guess it's back to two chronos, fortified with heavy acrylic glass, and lots of shots....
And that method is even within my own rather limited $-possibilities...! 👍🏼 😊
Matthias
-
I think the performance is at least as good as LabRadar claim on their website....
.177 Pellet 30 yards;
.177 BB 30 yards;
22 LR 60 yards;
223 60 yards;
270 70 yards;
308 80 to 100 yards;
9mm 130 yards;
40 S&W 130 yards;
45 ACP 130 yards;
500 S&W 130 yards;
12 gauge Slug 90 yards;
Paint Ball 50 yards;
Arrows 50 yards;
I'm not at all disappointed with the results Ron is getting.... I hope I can duplicate it.... Ron says the background and terrain is important....
Bob
-
To extract meaningful parameters (range, velocity, angle) from radar data one generally needs higher signal to noise ratios (SNR). The higher the SNR, the lower the standard deviation of the parameter. So for low SNRs (like 20 [13dB]) the uncertainty of the measurement can be high. The mean value of the measurement is generally correct.
The standard deviation of the velocity is given by
V(rms) = lambda/(2T*sqrt(2*SNR)),
where lambda = c/fc, c is the speed of light, fc is the radar carrier frequency, T is the total measurement time, and the SNR is the ratio of the power of the signal to the power of the noise.
I don't know the carrier frequency of the Labradar, but let's assume it is 10.525 GHz (X-band) for the moment. So lets plug in some numbers. Just guessing these parameters, but, let lambda = 3e8/10.525e9 = 2.85e-2 [meters], T=10e-3 [seconds] (10ms), SNR=20 [unitless], then
Velocity(rms) = 2.85e-3/(2*10e-3*sqrt(20)) = 0.32 [meters/sec].
That's not that bad, about 1.04 fps. That's the standard deviation of the velocity for a look time of 10ms. At a pellet velocity of 300 m/sec, as an example, the pellet travels 3m. If one wants to know the velocity in a particular range that is finer than 3m, the dwell time is less, and the standard deviation of the measurement increases. If we wish to know the velocity within a 98.6% certainty, than all we can say is the velocity is between plus or minus 3* Vrms. (plus or minus 3 sigma). If that's good enough, then great!
When I designed radars, I tried to use a minimum SNR of at least 20dB [100] as my goal. This usually resulted in a lot of work for myself, as it is a challenge to get high SNRs for little (low radar cross section) targets. But using 20dB allowed me single look parameter estimation at a low standard deviation.
Take away is, single low SNR measurements are not good for calculating BC or anything else. The only way to get around this is to 1) improve the SNR of the radar (usually not easy) or 2) take multiple measurements to reduce the measurement uncertainty. SNR improvements can be made by increasing radar transmit power, increasing the size of the antenna, reducing the range of observation, increasing the target size, improving the receiver noise figure, lower loss signal processing or increasing observation time. Practically speaking, it's hard to improve, especially if one doesn't have internal access to the radar. The only thing a user could do is take multiple measurements and average them.
So Bob, your colleagues observations are consistent with theory.
One last thing, SNR declines as 1/R^4 for a point target. SNR gets really small really quickly. For every doubling of the range the SNR declines by 12dB (a factor of 16) Also there is such a thing as multipath, which can be constructive, or destructive. Multipath can make some target ranges be undetectable, while further ranges are ok. If one is bothered by multipath (and the target should be detectable at that range) usually a small change in height of the target (or the radar) can allow one to detect at that range.
-
Thanks, I got a part of that, anyways.... ???
Bob
-
Sorry to overcomplicate that... Hah, this is how I feel sometimes when you are explaining stuff! ???
Takeaway is the measurement isn't exact, it's an estimate, which is a random number with a mean value that is extremely close to the actual value. The standard deviation of the measurement (the random part) is normally distributed about the mean (bell curve). Low SNR measurements have higher standard deviation (wider and lower bell curve). To accurately estimate the mean we need to either take a lot of measurements and take the average of the measurements, or have a high SNR measurement whose standard deviation is tiny enough (narrow and tall bell curve) to ignore.
So when you mentioned the measurements showing velocity occasionally increased as the range increased what was observed was the random component of the measurement being large. (The SNR was poor at that range.) The random component was large enough, and on the wrong side of the distribution which made it look non-causal. Because, as we all know, for a non-actively propelled pellet, (out of the muzzle) the velocity is always decreasing. If you take enough measurements and take the mean, the value will converge towards the true value. If you took 10 measurements and averaged the results, then the standard deviation of the final result would reduce by a factor of roughly 3. Or 100 measurements, then the standard deviation would reduce by a factor of 10. It's no fun, but that's the only option one has to get high quality (low error) measurements without modifying the radar. Then one has to worry about the repeatability of the shots themselves.
Random numbers and statistics can be difficult to digest. Parameter estimation aka measurements have errors if one looks closely enough. The errors, or more accurately, the standard deviation of (velocity) errors decrease with increasing SNR as 1/sqrt(SNR) as the original equation showed.
It's kind of like machining, nothing is really flat or smooth if one zooms in enough. Even a surface plate. Got a big enough magnifier, nothing is flat. You could measure a single tiny point that was out of whack and still meet the rms smoothness specification.
-
Here is a composite of 10 shots generated this morning along with a sample of
extended data readout (with processing). Below Mn .63 (SNR 24, dist 200 ft) the
Cd curves begin to scatter and by Mn .6 (SNR 21, dist 240 ft) things have really
loosened up. In order to get good data further down range, the unit was moved
out 50 yards (shielded with a piece of 2x12) where it can be controlled with a
smart phone (app from LabRadar website). The average of the second set of shots
has been added to the composite, the red dashed line labeled "extended". Very
little blending (< 1%change) was required to mesh it with the first set average.
Bruce, the unit operates at 24 Ghz, actual transmitted power is only 3mW. The
readout is on a two millisecond interval but I have no idea of the sample time (your T).
Perhaps you could make an educated guess.
Ron
-
Ron, thanks for posting that, it makes the problem perfectly clear.... Does the smart phone "App" require connection to a phone service, or does it work via Bluetooth?.... I have an Android tablet, but we have no cell service out here....
Bob
-
Bluetooth, so your tablet should do the job provided
it's at least version 4.4 .
-
Ron, interesting stuff. Is the SNR in dB, or linear? Theory predicts the SNR will reduce as 40*log10(200/240) = -3dB. (This is due to the 1/R4 in the radar range equation.) The simple equation is valid, if the target is a point, and if there is no multipath (reflections off the ground or sides of buildings, etc.). If your values for SNR are in dB, a drop of 3dB makes sense. If not, either there is multipath happening, or something else. Without any other knowledge of your setup (pictures of your test, and the environment) I can't even hazard a guess.
Pardon me, but I don't know the parameters that are in the graph. Cd is the drag coefficient? Mn is what? It's my turn to be puzzled ;)
Going back to the equation that I originally posted, we now have fc = 24GHz, so lambda = c/fc = 3e8/24e9 = 1.25e-2 m (1.25cm or about 1/2"). If the update time is 2ms, T has to be less than or equal to 2ms :). So let's see what T=2ms gives us. I will use the MKS system of units to make our lives easier. [meters, seconds, etc.]
V(rms) = lambda/(2*T*sqrt(2*SNR)) Eqn(1)
Since I don't know if SNR is in dB or not, I will use linear. (Kind of like a worst case...) SNR = 21 or 24
V(rms) = 1.25e-2/(2*2e-3*sqrt(2*21)) = 0.482 m/sec for SNR = 21 (1.58fps) 1 sigma
= 1.25e-2/(2*2e-3*sqrt(2*24)) = 0.451 m/sec for SNR = 24 (1.48fps) 1 sigma
So let's assume that most of the time, the reports from the radar come back within +/-3 sigma of the mean value. For a bell curve, that would be 99.7% of the time. (The reports would be within +/-2 sigma, 95% of the time, and +/-1 sigma 68% of the time) https://en.wikipedia.org/wiki/68%E2%80%9395%E2%80%9399.7_rule (https://en.wikipedia.org/wiki/68%E2%80%9395%E2%80%9399.7_rule) Another way of looking at this is 4.2% of the time, the report could be from 2-3 sigma from the mean. On average, the reading could be "off" from the actual value by 3-5 fps, 4% of the time. I'm not sure that these kinds of errors matter for your measurements or math, but it's good to be aware of these potential sources of error. Do you have plots of the velocity vs distance that you can share? I have no idea what the velocities are at 200 and 240 feet. If they are within 5-20 fps of each other, the measurement errors (1 sigma, etc.) could be significant. If they are more like 100fps different, then the errors wouldn't be significant.
If T = 0.2ms, then the errors would be 10x greater! I'm guessing that Labradar is doing it the right way, and using the whole observation interval that is available.
Let me say that so far, from what I've heard reported about Labradar, I continue to be impressed. It looks to be quite capable.
-
Bruce, to help answer some of your questions.......
- get a copy of the users manual, it has specs and general info about the unit
mylabradar.com/download
- click on the sample shot text file, most have overlooked it (only downloaded once)
- Cd is drag coefficient, Mn is Mach number.
Ron
-
Ron, just downloaded the User Manual and the Quick Setup Guide. Can't seem to find the sample shot text file on the download page. It could be right in front of my face... Can you humor me and point it out? I see two Firmware downloads, the User Manual & Quick Start Guide.
From what little I've looked at, Labradar outputs velocity at selected user programmable ranges. How many ranges are available? I presume you are doing the calculations for Cd and Mn. Is that correct?
Thinking out loud here; at 20C, the speed of sound is 343 m/s. (Had to look that up.) You have Mn = 0.6, give or take, and that corresponds to 205.8 m/s (675 fps)? What I can't extract yet is the distance information, or the velocity vs distance.
How are you determining Cd? Is from a differential velocity change, or some other way? This isn't my area of expertise, so if you can, please point me to where I can learn more about this. Thanks.
-
Answering one of my questions: How many ranges can Labradar report on.
From the FAQ: LabRadar will report the muzzle velocity and up to five other velocities at the distances that you select in increments of one (1). You will be able to select these distances prior to shooting. You can select feet, yards, meters, etc. Velocities can also be reported in metric and standard values.
Does this mean, you can get velocities at 200,201,202,203,204 feet ? Is this what is used to estimate Cd?
-
Found a paper on using radar for determining BC. https://arxiv.org/ftp/arxiv/papers/1608/1608.06500.pdf (https://arxiv.org/ftp/arxiv/papers/1608/1608.06500.pdf) Downloaded it, now I have to digest it. Unfortunately, all it says is that it can be done, but does not discuss the methodology.
-
Bruce, the sample shot file is an attachment to my post, just above the graph,
easily overlooked. It's what we refer to as extended data and contains 200+
points. Bob and I don't use the five programmable readouts from the radar
screen, they're entirely inadequate for our work.
In a nutshell, Cd is determined by relating drag force to velocity loss. I use
a slightly different equation than the Courtneys.
One other thing......BC and Cd are two entirely different animals.
Ron
-
Yes! The file was in front of my face! I have it now. The data is much more comprehensive than 5 values. OK, I need to plot some stuff to understand things. Been trying to come up to speed on this. Lot's to read. Back with some plots...
Units are seconds, fps, feet, unknown, dimensionless, dimensionless correct?
-
The LabRadar generates two different kinds of reports and stores them on an SD card so that you can transfer them to a computer Excel file.... The "preset" distances are also reported on the screen, for example you can have it report the velocity at 0, 10,25, 50, 75 and 100 yards.... Those are also saved to one type of report.... The other report is much more detailed, and looks like this....
(https://i378.photobucket.com/albums/oo221/rsterne/LabRadar%20Data/LabRadar%20Sample_zpsstvodnv2.jpg) (https://s378.photobucket.com/user/rsterne/media/LabRadar%20Data/LabRadar%20Sample_zpsstvodnv2.jpg.html)
You can change the units in the first 3 columns to what you want (eg. m/s, yards, metres, etc)
. Note that once the bullet is acquired the data is reported every 2 mSec.
If you plot the velocity and range, you can get something like this....
(https://i378.photobucket.com/albums/oo221/rsterne/LabRadar%20Data/25%20gr%20JSB_zpsqjgsyz98.jpg) (https://s378.photobucket.com/user/rsterne/media/LabRadar%20Data/25%20gr%20JSB_zpsqjgsyz98.jpg.html)
The equations are for the trendlines generated by Excel, but we are finding that doing that too early in the process makes things worse, not better.... The Cd is calculated by the following equation....
Cd = -2*Mass*LN(V2/V1)/(Rho/32.174)/Area/(D2-D1)
Mass = bullet mass (slugs)
V2 = far velocity (fps)
V1 = near velocity (fps)
Rho = air density (lb/CF)
Area = bullet frontal area (sq.ft)
D2 = far range (ft.)
D1 = near range (ft.)
If you do this for each pair of data points, you can then plot the Cd vs the mach number.... The mach number is calculated by dividing the average velocity (V2+V1)/2 by the speed of sound at the ambient temperature.... You can do this for 10 shots, and then plot all the Cd/Mn pairs on a graph and let Excel find the trendline, like this....
(https://i378.photobucket.com/albums/oo221/rsterne/LabRadar%20Data/LabRadar%20Cd_zpsuofctyel.jpg) (https://s378.photobucket.com/user/rsterne/media/LabRadar%20Data/LabRadar%20Cd_zpsuofctyel.jpg.html)
In this particular set of data, for an 8.4 gr. pellet in .177 cal, the velocity dropped to Mach 0.65 (730 fps) at about 50 yards, and as you can see the Cd spread becomes huge below that.... Please note these three charts are all taken from different data sets, and are just examples.... Ron's plots are for each bullet, rather than one chart for all 10 shots.... There are other methods to come up with the Cd curve as well, some of which are "manual/visual" in nature....
Bob
-
Bob, thanks for posting. That's helpful. I like the detailed data a lot better! Ok, as a radar guy, I like to start out with some simple plots - just to validate I can think about this the right way. Here are 2 plots, which aren't useful for you directly, but help me understand what is happening. Velocity vs Range and SNR vs Range. I hope the plots show up. I can't see them in the preview.
The SNR plot tells me the 'units' or SNR are dB. This is because the SNR (for a point target) should drop about 12 dB per doubling of the range. If we look at 50 feet the SNR is about 43dB. So at 100 feet the SNR should be about 43-12= 31 dB. From the plot its about 34 dB. That's pretty close for experimental noisy data. Likewise if one compares 100 and 200 feet, it's about 12 dB lower.
-
Bob, when you say for each pair of points, do you mean adjacent data points on the curve? So for instance you difference 41.7 ft and 41.29 ft and then 41.55 ft and 41.29 ft?
-
That is correct.... one data pair is one time increment, ie the Cd is calculated every 2 mSec.
V1 and D1 are at "X" seconds, and V2 and D2 are at "X + 0.002" seconds....
Bob
-
For this equation:
Cd = -2*Mass*LN(V2/V1)/(Rho/32.174)/Area/(D2-D1)what is 32.174? Is that g the acceleration due to gravity? So g has units of ft/sec2? Is LN the natural logarithm?
For Ron's data (the only complete data set I have), what is the mass of the projectile? Please include the units, be they grains, grams, or slugs. Also the projectile frontal area would be helpful. As well as the air temperature. I'd like to play around with the data and knowing this would help. One thing I did note was that the velocity is significantly noisier past 125 feet. I plotted the range and velocity difference (adjacent points) The range difference stays at the same noise level, but the velocity difference clearly gets worse with increasing range (or time).
Likewise if one does the division, V2/V1, the behavior is the same.
As well as ln(V2/V1)/(D2-D1)
-
One could apply a median filter to the data to knock down some of the fliers. It seems to improve performance out to about 250 feet.
-
Yes, 32.174 ft/sec^2 is the value for "g" to convert lbs. to slugs for the air density.... and LN is the natural logarithm....
You can clearly see what the problem is from your plots.... Past a certain range (ie below a certain SNR) you start getting V2 > V1, which is impossible.... This yields a negative Cd.... This is almost always followed by a large swing in the other direction, and the two extreme values tend to average out.... Of greater concern is what Ron showed in his first chart, where the Cd tends to increase drastically below about Mach 0.65 with the LR unit at the muzzle.... However, moving the LR out to 50 yards totally changes the Cd values to what is expected....
I don't know what a "Median Filter" is.... ???
Incidently, here is the same chart (all the data points for 10 shots, .177 cal) as the bottom one in Reply #16, but with the vertical scale changed....
(https://i378.photobucket.com/albums/oo221/rsterne/LabRadar%20Data/LabRadar%20Cd2_zpsu3dfkdsu.jpg) (https://s378.photobucket.com/user/rsterne/media/LabRadar%20Data/LabRadar%20Cd2_zpsu3dfkdsu.jpg.html)
This is more typical of what we might expect for a Cd curve for a pellet.... Note that Excel uses all the data points (as shown in the previous chart) to calculate the trendline equation (it is the same), even though some of the points are not shown because they exceed the vertical scale limits....
Bob
-
Of greater concern is what Ron showed in his first chart, where the Cd tends to increase drastically below about Mach 0.65 with the LR unit at the muzzle.... However, moving the LR out to 50 yards totally changes the Cd values to what is expected....
I don't know what a "Median Filter" is.... ???
This kind of makes sense to me. What is happening is that the SNR is too low for these lower velocities. The low velocities are occuring at the furthest distances from the LR. Parameter estimation error dramatically changes with SNR loss. If the LR is moved down range, and the measurement done again, the longer range segments, with their lower velocities will now have higher SNR. This is because the LR is physically closer to the projectile.
The power received by the LR is given by:
Pr= PtGtGrlambda2sigma / ( (4pi)3R4)
The noise in the LR is given by
Pn = F kT B
The SNR is just Pr/Pn. So SNR is proportional to 1/R4. The LR doesn't care where the shooter is, only the relative position of the projectile to the radar. So moving the radar down range from the shooter will result in better SNR for the lower velocities (at longer range from the muzzle).
Median filters are just another tool in a signal processing engineers toolbox. A brief explanation can be found at https://en.wikipedia.org/wiki/Median_filter (https://en.wikipedia.org/wiki/Median_filter). It is commonly used to remove salt and pepper noise in images. I have found it can be useful to reduce impulsive noise. The median of a sequence is the value when 50% of the sequence is above the that value. Some simple examples. x = {1,2,3}, med(x) = 2. x = {1,3,2,5,7,4,9}, med(x) = 4. You see the order doesn't matter. In the last case, there are 7 elements. If we reordered them we would have y = {1,2,3,4,5,7,9}. 4 is the element which is in the middle, there are 3 elements above it and 3 below it. 4 is the median. The median filter I used is 15 elements long. This filter slides along the sequence. The first output is the median of the first 15 elements of the sequence. The second output is the median of samples 2-16, the third output is the median of samples 3-17, etc.
I have an idea... Let me try something. Be back in a while.
-
In order to illustrate the effects that surroundings have on SNR, I've attached
three pics from work last summer. They're not to the same scale and I apologize for
that.......at the time no thought was given to the possibility of sharing them in the
future. The pellet is Crosman 10.5 gr at 900 ft/s. Numerous shots were made and
analyzed, and an example was saved from each scenario.
Pic1 represents what we would see from a normal setup........shooter at a bench with
the unit three feet from the ground and shooting parallel to the ground. In this instance,
as well as the following two, the ground is my front yard (flat). For this shot there was
a small shrub at 70 yards off to the side approximately 15 feet.
Pic2 is with the shooting bench on an elevated porch. The unit is now above the ground
7 feet with the pellet traveling parallel to the ground. Nothing else is in the radars
"field of view".
Pic takes it a step further. Still shooting from the porch, the aiming point is now a
small, nearly stationary cloud about five degrees above the horizon which effectively
places the ground further out of the "field of view". As you can see, each change makes
an improvement.
Ron
-
Thanks, Bruce.... So a Median filter is like a moving average, except it takes the "center" value, rather than the average value.... We have played with moving averages of various breadths (eg. MA=5, MA=10, MA=20), and it certainly smooths the data.... Unfortunately, you lose a number of data points from each end (from already scarce data)
. For example, if you are using a 10 point moving average of 200 data points, the first available MA would be points 1-10, then 2-11, etc. and the last being 191-200.... That reduces the number of plottable points by 5 at each end (190 points instead of 200)
. Doubling the MA to 20 points brings you down to 180 points.... The same thing would apply to a Median filter, would it not?.... (I see how you can fudge the data by repeating end windows, but that is probably what caused the sudden upturn at the end of your curve)
.
The moving average curves I have done look very similar to your Median filter curves.... I do see, however, that the Median filter would discard high and low data points.... They tend to occur near each other in opposing pairs, so a lot of the time a moving average "averages" them out, but not always, because at some point the sliding window for the MA picks up one (eg. a high) but not the other (eg. a low)
. The Median filter should discard both extremes all the time....
Bob
-
Bob, sounds like you understand it. Moving average can reduce your data size - it's the price one pays. A sliding median filter can have the same loss, but some people a lot smarter than me learned that if you loop your data (wrap it around) the penalty in 'bad' points is not as bad. As you noted, in this dataset, that caused the upturn. I did play around with the data again, and aggressively median filtered both the velocity and the range to smooth it and reprocessed the data. I came up with a pseudo Cd vs Mach #. (At the time I didn't have values for all the constants in the equation). I have to say the result is not satisfying. Disregard the y-axis, all that is important is the shape. You can see below Mach 0.6 things look ugly.
Unfortunately, (most of the time) we can't save all the data points. Another technique that might work (at the expense of end points) is what is called forward backward filtering. Matlab calls this filtfilt. Maybe I can get a better result. That will take a while, got to remember how to do it. (Need to search old files where I did it.)
-
Ron, something I just thought of. I seemed to remember somewhere (maybe in the xarchiv paper I linked?) that people had to play with the pointing angle (azimuth and elevation) to get the best SNR. The paper stated they had to first set up on large projectiles before they could measure smaller ones. They started with 9mm. This is because real packaged antenna patterns are not as smooth as we would like them to be. They are in fact lumpy. I've measured automotive radar antennas at sensor level and found (depending on how good a job we did) the pattern deviated a bit from a cosn(theta) shape. The pattern ripple could be many dB (1-5dB). I have also seen evidence of squinted antennas where the peak of the pattern is not where we expect it to be. This all depends on how well the packaged system was 3D electromagnetically modeled and simulated. This is considered to be large scale modeling, servers need to be used for this. Since this product is considered to be 'consumer class', I don't know how detailed the simulations were. At some point when doing consumer goods one has to say, "that's good enough".
So the fact that minor changes in angle are affecting your results could be explained by the above. If I am interpreting your pics correctly, there is about a 4-5dB change in SNR due to the tilt of the antenna (elevation change).
Your pics do show no evidence of multipath, so that is good. Multipath would show up as huge dips in the SNR in a pseudo-periodic fashion (high frequency in close range, lower frequencies in further range).
I also don't see significant evidence of tumbling or large yaw. When I get a chance, I'll run the velocity data through an FFT to see if there is anything there. Nope. Nothing there.
-
Here is the filtfilt version of pseudo Cd and SNR vs Mach Number. Filtfilt smoothed out some of the trash, down to 0.6. Used an 8 pole Butterworth filter Wn=0.125. How much do I trust this? Wouldn't bet the farm on it. Seems like the data from 25 dB and higher is ok. From a practical standpoint, I'd recommend only retaining data with > 25dB SNR. The low SNR data just plain isn't dependable.
-
Ron, what is the weight and diameter of the projectile for the dataset that you posted? My Cd is off by a factor of 20. You seem to have numbers of ~ 0.4, I'm getting numbers like 8. ???
I guessed the weight was 8.5 grains, and diameter of 0.177"
Used a conversion factor of 0.000004440136 to change grains to slugs
For rho I have 0.002373 lb/ft^3
Computed frontal area simply by pi*((D^2)/2)
Is this sensible?
-
I read the Courtney paper while still trying to decide whether or not
to buy an LR. Their struggle with range vs projectile size was personally
discouraging but I had no such experience and was able to do much
better right out of the box. Since the half power beam width is only seven
degrees wide aiming is critical. I equipped mine with a mechanical sight.......
a short piece of tubing attached with Elmers. The unit was laid face down on
a surface plate and a small v block provided alignment until dry. Of course this
makes the assumption the radiation pattern isn't skewed from a line normal to
the case, as you suggest it could be.
In use, the gun and unit are always aimed at the same point. In my lay opinion,
a comparison of 2 and 3 demonstrate, more than anything, the reflectivity of
grass and the resulting degradation of SNR. I could, if you feel it worthwhile,
deliberately miss aim the two and see what happens.
For the sample shot...... use 15.7 grains, .22 cal, .002376 for rho, 58 F.......don't be
too concerned if you come up with slightly different numbers, there are a
variety of reasons.
Ron
-
LR might have made improvements since the paper was written.
Thanks for providing the HPBW. Is this the total HPBW, or just the transmitter HPBW? I haven't checked the specs yet.
Sometimes, as I noted, there can be a squint of the beam. I'm pleased that you have some sort of sight to help get consistent measurements. From what I see, just a few dB in SNR is greatly changing the results, (which is normal) so it might be worthwhile to check if there is any squint.
I'm not recommending this, as it's not intrinsically safe, but if you were in an unpopulated area and could point LR to the sky and shoot safely, it might be instructive to see if the SNR is improved. Please don't do this if you can't guarantee everyone's safety!!! It's kind of a rhetorical thought question. The experiment might answer the question if ground return actually is suppressing the SNR. It's not a perfect test, because the radiation leaks out of the back side of the unit, but it should give an indication.
I had the wrong diameter and weight. That might help get the values closer. Alas no, I have an error in my program. :( Hey, it's just a bug. It will be fixed.
How do you determine frontal area, just pi r2? Or something more sophisticated?
-Bruce
-
Bruce, not sure I think wrapping the data around for the Median filter is a good idea.... It doubles the number of points at each end, right?.... I would sooner use a set of 9, and then discard 4 datasets on each end.... Additionally, since the range data is much smoother than the velocity, would it be better to use a (small) moving average for the range and a median for the velocity?.... Just a thought....
Yeah, your Pseudo Cd and Filtfilt Cd curves are pretty much unusable.... BTW, are you watching your units?.... In Imperial units the frontal area was in sq.ft
. I would think in Metric you might need to use sq.metres?....
I don't think Ron's improvement in SNR is due to the angle change, I think it is simply a matter of removing the ground clutter.... To a lesser degree, that is why the 2nd dataset is better than the first, because it is higher above the ground....
Bob
-
Bob, the wrapping is done with just enough points to do the median filter. In other words, if the median filter is of length 15, only 15 samples were wrapped and used. The extra processing is minimal. No doubling of points. The range data is smoother than the velocity, perhaps a moving average would be ok. All the processing is being done on a PC, so it doesn't matters much. If this was to be implemented on an embedded platform, I'd think about stripping the processing to the bare essentials.
I thought I was watching my units! Everything was converted to Imperial, just to be consistent with your values. Apparently there is a problem - eventually it will get sorted out. I'll do a dimensional analysis, with all the units and see what comes up. Usually that's a very good way to resolve this sort of thing. Probably was a little sloppy somewhere, as the bulk of the code was written in a day. (About 170 lines)
My curve was derived from only 1 dataset, a single shot. That's all the data that is available to me right now. Using more data helps improve the SNR, which in turn improves the results. Using aggregate data can be beneficial.
Ron's improvement in SNR could come from a several places. The best way to find out is to construct an experiment or two to find out. A sky shot, if safely done, could eliminate ground clutter. If the LR was on a good tripod with accurate azimuth and elevation markings, and one did a sky shot, one could show if there is ripple or squint in the pattern. At the moment, we can't rule it out. To make it easier, one could scan only one dimension to see if it mattered, say in azimuth.
There's theory and there's data. Data (in my opinion) rules. Sure, there can be bad experiments, or technique, but once one has good technique, and a sensible experiment, data is reality. Given the choice, I'd rather measure things rather than guess.
-
I thought Ron's 3rd chart (actually first in the order shown) was a "sky shot", he said he was shooting at a cloud about 5 deg. above the horizon....
Bob
-
The antenna pattern isn't as abrupt as one might think. Say the beam is 7 degrees wide. What this really means is the half power beamwidth is 7 degrees. That means the power is down by 1/2, or 3dB. If the beam is pointed 5 degrees off the horizon into the sky, a significant part of the beam is still in the ground. Here is a figure to show what I mean.
Positive angles are toward the sky. Negative angles (including zero) are painting the ground. The red line is the half power points. The peak gain is 22dB. At the ground the antenna gain is 16 dB, or -6dB peak. There still is a lot of energy (clutter) coming in from the ground, especially from up close. If you recall (from my previous ramblings) the receive power falls as 1/R^4. If the radar is 3 feet off the ground, and pointed 5 degrees up from the horizon, the -6dB point hits the ground at 34 feet. r = 3 ft/ sin(5 degrees). The -10dB point hits the ground sooner. If you want insignificant amounts of ground clutter in the experiment, one needs to point the beam higher. As a SWAG, about 15 degrees up off the horizon or more. The pattern that was shown was idealized, the real one is worse.
-
Good to know.... so elevating the "Target" helps, how much needs to be determined by experiment.... Now all I need is a solid target backstop hanging 500 ft. below a blimp hovering at 1000 ft.... and never miss it, of course.... One problem is that gravity will be slowing the bullet if it is shot upwards, decreasing the velocity.... but I guess a correction could be done in the math to allow for that?....
Quick question, how much noise would you get from a forest of trees at 150 yards from the LR unit?.... How much would it degrade the SNR at 50 yards?.... 75 yards?.... 100 yards?.... Could I aim the LR at the top of the tallest tree, which is on a hill (say 15 deg. angle) and use that as a POA?.... How much better would that be than shooting level at 4 feet above a grassy field, also at 150 yards from the same forest?.... What would happen to the SNR if I backed up to 300 yards from the forest (noise level of 1/16th)?....
Bob
-
Hah, hah. What one really needs to worry about is what is called close in clutter. So one way of "fixing" this is to shoot over a valley or dip in the land. The ground should fall away, like a V. If the beam intercepts the ground at say 30 yards away, rather than 10 yards, the clutter power would be reduced by a factor around 10. That should be noticable. The target could be on the hill. Outdoor antenna ranges often are designed to have a dip in the land between the the Transmitter and the test antenna. Or, the Transmitter is put up in a tower and the test antenna is on its own tower.
A forest 150 yards away will have very little effect on the results. It's the stuff in close that can desensitize the radar.
Could I aim the LR at the top of the tallest tree, which is on a hill (say 15 deg. angle) and use that as a POA?.... Yes. Or shoot from hillock to hillock. One doesn't need a mountain or cliff. How much better would that be than shooting level at 4 feet above a grassy field, also at 150 yards from the same forest?.... Well, at least a dB, since that's what Ron's data shows. Beyond that, try measuring it. It's tough to estimate much more that this from a distance of thousands of miles away.
Getting the correct answer would take a lot of modeling and simulation, which at the moment, doesn't make sense to do. There's not that much public published papers on low angle ground clutter at 24 GHz, to my knowledge. Not that it's impossible, but it's not a casual effort, especially if one doesn't already have code in place to modify. Based on past experiences, it's like a couple thousand lines of code. That takes a while to write, and far longer, to validate. People have made careers out of this. Heck, at the moment, I'm not sure of what their waveform is. To do the modeling, the waveform is a prerequisite, as well as a terrain map. It's a lot faster to measure it than simulate.
That being said, I just found the LR FCC site, which does have User Manuals and a copy of their test report. https://fccid.io/PDG-LABRADAR (https://fccid.io/PDG-LABRADAR) Going to take a look there... Might learn something there.
-
Bruce, thanks for your help.... I'm not trying to create a make-work project here.... *LOL*
. Pardon some of the rookie questions, but when it comes to radar, I don't even know what I don't know....
Bob
-
Learned something from the internal pictures. The transmitter is a 4x8 patch array, it has about 22 dBi gain, according to the User Manual. LR is using a giant receiver array, it is 12x16. I'll estimate the gain in a little while. The chip that is in common to both the transmitter (TX) and receiver (RX) appears to have a similar footprint to the BGT24MTR, which is a 24 GHz transceiver chip. From looking up the part, I can confirm the small array is the TX. This makes sense, because the total TX power is capped according to regulatory agencies, like the FCC and IC. The RX array is connected to a low noise amplifier - which is good design, and then to the BGT24MTR. Now I understand how they get their sensitivity.
From basic antenna theory, we find the HPBW ~ 0.89 x lambda/D [radians], where lambda is the wavelength (c/fc), and D is the aperture. So the bigger the antenna, the higher the gain, and the narrower the beamwidth. In the LR, we see that the D_horizontal ~ (16-1)*n*lambda, and D_vertical ~ (12-1)*n*lambda. Typical antenna element spacing is a little wider than 0.5*lambda, but let's measure it since there is a scale on the drawing. It seems the actual spacing for the TX is about 0.72 lambda. If the spacing is greater, the beams will be narrower. So lets put in the correct values.
HPBW_RX_h = 0.89 * lambda/( 15*0.72*lambda) = 0.0824 rads = 4.7 degrees, give or take.
HPBW_RX_v = 0.89 * lambda/( 11*0.72*lambda) = 0.1124 rads = 6.4 degrees
I guess the design choice was made to have a narrower beam in azimuth than in elevation. As you can see, the RX beam is narrower than the TX. This does help, as the total gain is the product of the TX with the RX. So roughly speaking, the HPBW of the RX is about 1/2 of the transmitter in azimuth. This makes sense, since the array is twice the size. In elevation the beam is about 1/3 the width.
So... pointing angles or squint of the RX seem to be much more important than of the TX.
Also the specification of the antenna beamwidth is 7.6 degrees in azimuth, and 18.5 degrees in elevation, at least if I have the correct orientation of the RF board. So it seems the TX beam is painting a lot more of the ground than I expected. But it seems the limiting factor is the RX beam, and it's elevation beamwidth is about 7 degrees.
I'm still thinking that going from hillock to hillock is a good test. This was a good exercise.
Sorry, I can't quite post the image - it's 1.6MB, and the limit is 1.2. I tried. It's in the Internal Photos document at the fcc url.
-
Bruce, thanks for your help.... I'm not trying to create a make-work project here.... *LOL*
. Pardon some of the rookie questions, but when it comes to radar, I don't even know what I don't know....
Bob
No worries. I'm used to these sorts of questions. Please ask, it's how we all learn.
-
Here's a resized image, the original is on the FCC site.
-
Bruce, I sense you have a better grip on LR characteristics at this point
so can you outline a test that will expose the effects of "squint", "painted ground"
and "leakage out the back" and what it's costing in performance?
Some of us live on very flat ground and have no problem with sky shots.
Ron
-
Ron, I have to think about it a bit. Now that I know the basic characteristics of the beams, I'll come up with a plan. I suspect that 15-20 degrees in elevation would be sufficient, but need to do some calculations.
Still pondering how to suppress "leakage out the back" or anywhere else without using RAM (radar absorbing material). Maybe I can think of a poor man's version. Last time I had to buy some RAM, it was $55 per 24"x24" piece, and messy as all get out. Not recommending it. But... if you know of some black styrofoam, it might be carbon loaded and absorb 24GHz. You need at least 6-12" thick, half that if you glue on some aluminum foil on the back. Gee, I hope I won't have my black magic card revoked :-X
Based on your data, and the information from the FCC site, I've estimated the Labradar link budget. It helps me think about the problem and how the pattern would affect the SNR. In a day or two, I'll have a definitive plan.
-
Just out of curiousity I figured out the velocity loss due to gravity for a Time of Flight of 0.4 sec.... such as you might get with a pellet of BC = 0.035 at 950 fps.... when launched at a 15 deg. angle upwards over a 100 yard distance.... It works out to only 0.44 fps, which is below the accuracy of the LR unit, so angles up to that can be ignored, IMO....
Bob
-
I agree, Bob, the additional effect is pretty small.
And just thinking out loud, if push comes to shove we
can always rotate the unit 90 degrees in order to
minimize the "painted ground" part of this puzzle.
Ron
-
Agree with the both of you.
Got a chance to think about this some more. The actual elevation beam is narrower than the Transmitter beam alone. This is because the effective beam width is the product of the Tx beam with the Rx beam. So the composite elevation half power beam width is about 6 degrees. (It's basically dominated by the narrower beam, in this case the Rx.) If the antenna had low sidelobes (leakage in undesired directions that are not in the main beam) the gain would be down by 30 dB at 10 degrees below (and above) peak gain. So if the beam is at 10 degrees upward from the horizon, things should be mostly ok. (Of course this means the trajectory of the pellet has to be in the beam.)
You could try rotating it on its side as well, but I'm not sure it will matter. The sidelobes are most likely what's the dominant effect. There is no way (from here) to estimate the sidelobes of the antenna. That either has to be measured, or whole system 3D EM simulated. Impossible to SWAG. The only thing we have in our favor, is the energy of the sidelobes will be dependent on the product of the antenna gains as well. In this case, it helps. OK antennas have -20dB sidelobes. Great ones are -40dB.
Your call. I recommend a 10-15 degree skyshot and compare it with 0 degree elevation. Shots to be made on the same day, if possible. (Minimizes setup and environmental effects) Ideally, at least 8 shots of each. That way we can average out a little of the ordinary statistical variation. Hope this helps.
-
I have a question on the number of shots (sample size)
. Is it based on the ES or SD?.... In other words, if the ES or SD is larger, should we be sampling more shots?.... What about "visually" throwing out shots that don't seem to fit with the majority?.... You mention 8 shots, that is a perfectly feasible number, IMO.... What is the advantage of more?....
Bob
-
I generally use SD. As tempting as it may seem, (unless something is very wrong), I don't throw out shots. One needs to use judgement, however.
Cherry picking isn't a good way to see what is really happening. In a way you are injecting user bias, i.e. just picking what you want to see. It sure is tempting, though.
In general, the more data, the better. It gets awkward dealing with it, but that's what one has to do. To answer your question, the variance (square of the SD) is reduced linearly with the number of samples (N). So the SD reduces as the square root of N.
I chose 8, because I didn't want to scare anyone away ;) More is better. It soon becomes a data managing exercise. 16 datasets of the same experiment would have 1/4th the SD of 1 dataset if averaged over the ensemble. 100 would yield 1/10th the SD.
This is of course, assuming the same experiment. The pellet is in the beam, the shot was "good" and an o-ring or seal didn't blow. If the rifle fell off the bench and fired, don't count that one!
As an aside, when the datasets get large, I write programs to do the processing and reduction. Especially if there are a lot of files. It becomes a lot easier (later on, especially) if the data is handled programmatically.
-
Miles Morris, who is the ballistician who wrote the stability articles in the "Workshop", has 40+ years of experience in Exterior Ballistics (much of it using Doppler Radar).... He stated that the later you apply any averaging/processing the better, and that averaging the distance and velocity data tends to give unrealistic results.... He recommends using the raw data to get the Cd and then do the averaging at that point.... I'm not sure I am 100% convinced of that, but I am leaning in that direction.... The "uptick" in the Cd at low velocities seems to be worse if the data is averaged early in the calculations....
Bob
-
He's right.
The longer you can keep the data 'unmolested', the better and more accurate the end results. When one preprocesses the data (sometimes for our intellectual convenience) one can unwittingly insert small and subtle biases. The biases are errors which end up accumulating. If we carefully do preprocessing, it's possible to avoid some errors. It's far better from an accuracy perspective to use all the raw data to get the results.
For the example that you are using, the 'bad' values in Cd at low velocities, are due to an insufficiently sized data set (for the low velocities). The SNR is so low at these velocities, that the reports are noisy. To reduce the noise, one needs to increase the number of shots (experiments). It is only by averaging more experiments that the SNR (for these low velocities) can be made good enough. Or, as previously noted, by moving the LR downrange. to increase the SNR for the lower (formerly long range) velocities.
-
I have been corresponding with Ron Burnett, who is working with his LabRadar on trying to develop Drag Coefficient curves for pellets.... Ron has tested both .177 cal and .22 cal pellets, and after some distance downrange he is finding that the data becomes inconsistent.... He shoots over an open field with no obstructions, and a long distance behind the target before any trees.... He has determined that when the Signal-to-Noise Ratio (SNR) drops below about 20 is when the data becomes inconsistent.... In his setting, that is at about 50-55 yards for .177 pellets and at about 75 yards for .22 cal.... even though there is a radar return at much longer ranges....
Bob
I know I'm resurrecting an old thread, but it seems appropriate. If I'm wrong feel free to delete it and I'll create a new one.
I have only had my labradar for just shy of 3 weeks now and use a recoil trigger, actually built the trigger before the labradar arrived. My normal shooting lane is 55 yards and works best on low power of my labradar due to interferences and SNR.
Have you experimented with making small adjustments in your aimpoint of the air rifle while leaving the labradar and shooting position fixed? I just yesterday was going to set up a better aiming solution on my labradar and started by setting it up about 10 feet above ground level shooting off my deck, aiming at the top of a tree at 114 yards and firing a series of shots to determine where the radar was truly aimed. I was quite surprised to see how much better SNR can get with just a 3+ foot different point of aim of my Uragan at 114 yards on low power. I didn't try it on high power as my normal shooting spot doesn't like that. After testing I need to come up with an extremely accurate method of aiming my unit.
My results, left out all the wasted shots that were worse. That is a huge difference for just 3+ feet in difference of rifles point of aim. You see on low power it only tracks 22's out to 55 or so yards. Open the image in a new tab to see full size.
(https://i.imgur.com/c0r3ouZ.jpg)
added for clarity: I did try shooting just a little further out on the line where I had best SNR and the SNR drops very quickly going out from the yellow circle.
-
Wow, that's very interesting!
Thanks for posting your data and the perspective that this opens up!
Matthias
-
Interesting to see how short the LabRadar range is on low power.... The SNR degrades a lot, even on the good .22 cal track, past 27 yards.... :o
The LabRadar is very sensitive to backscatter from the ground, and ultimately from the background (in your case forest).... I changed from shooting over a level field to angling my shots across a small valley (about a 7-10 foot dip), shooting from a rise on one side to a 100 yd. target on a rise on the other.... and it made a significant to the consistency of the readings.... The caliber makes a big difference as well, of course....
Thanks for the post, and keep us advised of your findings....
Bob
-
That's not forest, my 55 yard target is 20 yards deep in the forest shooting through an opening no more than 12 feet accross and 7 feet above ground level. Shooting downhill and 20 yards from shooting position the bullet path gets within 1.5 feet of the ground before ground falls away. It's why I use the labradar on low power. I bought it after my old chrono started getting flaky. The manufacturer looked at pictures of my shooting lane with yardages of everything and said it would work fine for my use, which was strictly load development on firearms and accurate close in data is all I cared about. The labradar has actually tracked some 30 cal stuff to within a couple yards of my 55 yard target, with lousy data of course. Actually gives good data right to water jugs at 18 yards when I was testing out a few 30 cal subsonic bullets for hunting.
Where I shot the test is the most open place on my property, and I would never shoot a firearm there. I was shooting 9.3 gr 22 pellets for that test, the highest shots probably landed just inside my neighbors property which on that path is only 144 yards from where I shot from.
-
Before anyone reads this and it reinforces your opinion the labradar s000ks, I had many emails back and forth before purchasing with the manufacturer showing them pictures of my two different shooting lanes I use at home with yardages of everything, topographic details, and anything else I could think of and they asked for. I KNEW going in I would have no good long range data on my home range, I just needed short range GOOD, RELIABLE, AND ACCURATE data for load development on firearms. They assured me on low power it would work for that, and it absolutely does far more accurately and consistent every single time versus optical chrono's I've used. My last one was getting super flaky which is why I looked into the labradar and am happy I did.
Interesting to see how short the LabRadar range is on low power.... The SNR degrades a lot, even on the good .22 cal track, past 27 yards.... :o
The LabRadar is very sensitive to backscatter from the ground,....
Bob
Actually, I just today discovered it was maxing out at 55 yards no matter what. It turns out that when you set the 5 individual distances it reports in the series(not the tracks), it will only track 10 yards past whatever the longest distance you put in. I had my 5th reporting yardage set at 45 yards, changed it to 105 yards and bingo. It actually gave readings today both on low and high power out to pellet impact at my 110 yard range, even on low power. However, it reported impossible distances and SNR was single digits much past 50 yards, I saw the pellet impacts in the red clay bank which is soft from rain right now, it reported one pellet out to 115 yards and two more to 114 and several at 113 yards, that were all in a 8 or so inch group in the clay, the pellets were impacting and burying in the clay 1/2 yard past my lasered 110 yard target and sending up a small amount of spray. I suppose my laser rangefinder could be that far off, but I seriously doubt it. Any data returned was worthless past 22 yards (data is actually better on low power with the background interference I have), my 110 yard range has too much interference in background starting at about 19 yards, data is strangely better when shooting at my 55 yard target that is shooting straight into the woods with bullet path getting closer than 2 feet to the ground at 20 yards.
Pulled out my 300 blackout pistol to shoot a few boattail 220 ballistic tips to get rid of them, bad loads performance wise and the nosler 220 ballistic tip specifically for the blackout subsonic hunting(suppressors are wonderful things) is way to fragile to pull and reuse. The back of the boattail is larger than the .22 pellet skirt, but the manufacturer tells you boattails present issues for the labradar and the tracks were worse than the 22 pellets, but in both cases garbage velocity past 20 yards on my 110 yard range on low power and high power was sooner. The labradar was reporting to 111.9 yards on the blackout which coincidentally, my laser said 112 yards on the clay bank where the bullets were impacting, all velocity data was garbage past 14 ish yards on high power with the blackout, it does a few yards better on low power. This is what too much clutter/interference/signal backsplatter looks like when you graph it. It looks like two different graphs, it is the exact same data, just changed the X and Y axis minimum and maximum to keep the graph from being so compressed and not looking all that bad until you get out some distance. With all the interference I have, when the SNR gets down to 28 the velocity reading are garbage when shooting on my two ranges. This graph is on the high power setting.
(https://i.imgur.com/6fgt9qt.png)
-
Karl,
thanks a lot for posting, I appreciate this kind of obscure information for the obscure parts of our hobby/obsession. 👍🏼
Since I really would like to know the BC of my gun's favorite projectiles out to 100 yards.... what I gather is that I should save myself the hundreds of dollars for a LabRadar
and instead get for my money a second normal chrono, a 1" safety glass, and a whole bunch of ammo (since I'd really need to nail those shots over the chrono sensors at 100y).....
Am I understanding this correctly?
Matthias
PS: I might have access to a long factory building for my BC tests, but all this talk about background noise will probably disqualify that kind of environment for a LabRadar.....
-
If the caliber is large enough, you may be able to get good data with a LabRadar out to 100 yards in ideal conditions.... ie shooting across a valley, or up on an angle towards an aiming point on an isolated mountain, where you know the pellet will land where nobody is.... eg. across a lake.... I have my LabRadar set to report the velocities at 30, 50, 70 and 90 yards, and I have my targets at 100, with a low area of field in between.... It works pretty well with a .35 cal, marginal with a .30 cal, and pretty solid to 70 yards with a .25 cal, but only to 50 yards with .17 cal.... I haven't tried elevating my aim point as of yet, but can likely do so in safety.... However, I would not be able to simultaneously check the 100 yard accuracy if I do that....
Bob
-
Yeah, Bob, thanks for that info.
That's what I was afraid of, of course, that smaller calibers aren't so suitable.
Because .22cal is really "my caliber" as my resources would best be served by an allround do-it-all caliber....
Now, I do have vast desert areas and tall desert hills right behind my house where I could shoot to my heart's content and to the LabRadar's content as well...! 😊
However, in those hills the wind HOWLS like a pack of wolves year around. ::) So, no accuracy tests there.... .
🔶 Bob, just to mention this.... 🔷
I have not given up on the DREAM that you and some of us were dreaming just a few years back....
The dream that you would improve our drag model for pellets...
maybe helping us move from a GA model to a GB model (with the B standing for Bob, or maybe GBM, for Bob and Miles)! 😄
I'll attach the logo I made back then I checked the date, it was the 3rd of October 2018.... 😄
Even if this project might not be possible:
🔶 ➔ I am very very thankful to you for all that you have done for me and COUNTLESS OTHERS in SO MANY YEARS of researching, testing, writing, and advising! 😊
Matthias
-
Karl,
Since I really would like to know the BC of my gun's favorite projectiles out to 100 yards....
Am I understanding this correctly?
Matthias
.....
I would say no, you are not thinking it through. I've been shooting with optical chronos for well over 30 years, they all have issues and none are as accurate and repeatable as the labradar.
With the labradar, you just need to shoot the cheapest pellets or whatever and find out how to aim the labradar for best SNR and in the process looking at the tracks find out what your location is good for in terms of range and good data. On mine the radar array is definetely not facing perpendicular to face of the unit, not saying anything bad about that, just the cost of not spending 10,000+ dollars on a doppler radar unit if you could get someone to sell it to you(I'm sure there are at minimum severe licensing issues with the government). The labradar is absolutely accurate as long as it gets a good return signal vs backscatter, in my case with all sorts of "bad" terrain/trees/metal/etc.. that means 20 yards max at low power, in your case it may mean 50 yards at high power or even more than that. You have to test it. The labradar will trigger on doppler setting, on the doppler setting it waits to sense a projectile flying in front of it, better have it running on a li ion battery pack when doing that. So you could work out BC's from your shooting position to whatever the max range of good data you get. Then move the labradar out to 10 or 15 yards short of that distance and set it on doppler trigger, to get good data from that point. Keep doing that until you get out to the distance you want. BC is not a constant, it changes with velocity, published numbers are an average in whatever velocity range the publisher/manufacturer is using. If you are shooting on a public range, good luck with putting your expensive labradar downrange, I would not do it personally.
They can be used indoors, depending on construction you may only be able to get 10 yards of good data or you may be able to get way better than that, if you don't test it you don't know. It is specific to the environment you shoot in. I believe, but don't know, the labradar would have to be operated on low power indoors.
One other thing, reading the years of posts on the net about the labradar, one thing is clear. Triggering the unit can be a pain, some people just b#$($ch and moan about it, others contact the manufacturer who is good on support and work out their problems (emphasis on the persons usage, not the units problems). And sometimes the manufacturer has you send it back and they make it right. Seems the microphones on them are not uniform and some are not up to the task, another cost of not spending a lot more on the unit, but the manufacturer will make it right if that is a problem. The answer of course is to just use a recoil trigger from the beginning, never a problem with them. I ordered the parts I needed to make my own recoil trigger the same day I ordered the labradar. Other than the first day my labradar showed up, I have never used the internal microphone trigger. I played with it out of curiosity within half an hour of the labradar arriving. It didn't take me 5 minutes to be glad I had the recoil trigger parts on order and the last thing I needed arrived in the mail the next day. Going through an entire box of 50 high velocity 22lr, shooting a 16" bolt action rifle, once I found the best placement of the muzzle to the labradar unit, it would trigger 4/5 times. It never missed a shot accoustically with a 41 magnum desert eagle and didn't care where my muzzle was. I have never missed a shot with the recoil trigger I made. I have forgotten to arm the unit, taken too much time between shots and didn't pay attention to the unit disarming(I have my disarm time set to 1 minute, saves power and not big deal to press a button to rearm), and a couple times shooting a string I did not look at the unit between shots and had gotten a failure to track error that you must respond to. My failures to track have all been me causing enough disturbance to the rifle/pistol that the recoil trigger signalled a shot.
In short, if the price of a labradar doesn't hurt your feelings and you don't have unrealistic expectations, it is the best thing since sliced bread. Beats the pants off any optical chrono ever made.
-
I forgot to add, on caliber, in my situation/environment, I shoot my 17hmr and it tracks them just fine. Maybe 1 yard shorter on average for good data, but it is for all intents and purposes in my personal shooting environment no different than tracking a 44 magnum. I'm sure if I had a better range to shoot on there would be noticeable differences, but muzzle to 20 yards no difference for me from 17 cal to 45 cal. In my environment anything I track over 20 yards data is questionable, doesn't matter that it can track 22's out to 100 yards, the data is horrible and useless. In your shooting environment you have to find out what that limit is.
-
The last 2 posts address one of the problems we airgunners face, and with slugs as well, not just with pellets.... The Drag Models we use (GA or G1) do not match our projectiles very well, which is why the BC varies with velocity.... If they were a perfect match, it would not, the BC would remain the same, regardless of MV....
I had great ideas of shooting tens of thousands of pellets and slugs, using the LabRadar to track them, changing the velocity (or range), repeating, and the "stitching" the data together to find those elusive drag models.... I thought that once I retired I would have lots of time, drive and energy to do that.... Unfortunately that has not been the case, and I seem to have less time to devote to airguns, rather than more.... Add to that the LabRadar being fussy in terms of background interference, I haven't even begun my goal, let alone achieved it.... :'( :-[
Bob
-
Bob,
life sometimes goes a bit different than expected. I'm sorry. Please, hang in there, and keep shooting as long as you can!
And keep writing!!! 😊
🔹As I take care of ageing parents I'm wondering about my own age and what will come of my airgun obsession when I'll get there.... 🤔
Will I be able to stalk quarry? Or at least walk at all?
Will I be able to spot quarry? Or see at least a big bull's-eye?
Will I be able to lighty squeeze a trigger? Or at least hold a spoon?
I trust my God has a plan for my last years and even though guns in heaven seem a somewhat screwed-up interpretation of it I'm sure it's gonna be even better than hitting PD's at 300 yards with an M3 shooting 70 grain BBT's...! 😄
🔹About the LabRadar: Karl, I appreciate you for clarifying the issues, and the method that I would need to use to do BC / drag model testing.
Now, I do take issue with one of the things you said at the end of your detailed post: 😟
That the "LabRadar is the best thing since the invention of sliced bread."
➔ Wow, that's a really really unhelpful phrase to use with me as I'm trying to be responsible with my finances...! 🤣
Seriously, kidding aside, Karl, thanks. 👍🏼
I admit to have been looking in the classifieds for a second optical chrono at a good price, but they are rarer than rain in the desert region where I live....
Matthias
-
This conversation has stopped for a few years but I would like to share a little of my experience with the Labradar. I would like to tell you that I think it is a fabulous tool, I have managed to have shot tracking up to 140 m (153yards) shooting caliber projectiles in an open field. 25. One of the points that no one commented on was how they used shot detection. Doppler trigger detection does not have good precision, so it is advisable to use an inertial or acoustic trigger.
I made my acoustic trigger and compared it to the doppler results the distance determination with the doppler trigger varies and is not consistent. You can see the manufacturer's manuals where this is expressly indicated. Consistent results will not be achieved when using a Doppler trigger to prolong a shot with the radar positioned further forward.
My goal is to be able to develop an improved plow curve for Slugs. The procedure I am using is to make shots, model them individually and then obtain the average in order to reduce the error in taking speeds. I attach some graphs of speeds and distance obtained by my Labradar.
-
I use the LabRadar microphone to trigger the unit, and I have it extended a bit (8-10") in front of the unit, about level with the muzzle.... Getting a signal out past 100 yards is possible, but when one velocity is higher than the preceeding one (which is impossible, of course) that limits the reliable range of the data.... You can see that happening with your lower chart at about 40 m, and it becomes significant at just under 90 m.... There are also low spikes in the SNR about the same distances, and past 90m....
Bob
-
I'm glad this thread is not dead. 👍🏼😊I'm still very interested in BC testing.
I just don't have many opportunities to go out for a day of shooting and now I lost my 100 yard (cow corral) range....
The Labradars dropped in price: used units on Snipers' Hide are going for 300 and 400$.
Matthias
-
I'm glad this thread is not dead. 👍🏼😊I'm still very interested in BC testing.
I just don't have many opportunities to go out for a day of shooting and now I lost my 100 yard (cow corral) range....
The Labradars dropped in price: used units on Snipers' Hide are going for 300 and 400$.
Matthias
As @rsterne Bob mentioned, it is better to use the LabRadar microphone to trigger it. This better synchronizes the radar emissions with the actual shot timing. And that means better quality radar measurements (higher signal to noise ratio). If you are unsynchronized to the shot, the radar can be looking for targets before the shot. This translates to a loss in target energy. In the radar business, this is called eclipsing loss. Radar engineers (that was me before I retired) wish to minimize radar system level losses, to get better performance.
If the Labradar was not mishandled, (dropped or otherwise damaged,) a used one should be fine. If I had a place to set one up (a nice long range) I'd buy a LabRadar. Now, these radars aren't perfect, but they give us access to information that would be rather hard to collect any other way.
-
Without a doubt we cannot have speeds that increase with distance, we must consider that every measurement has an error and that error in the case of the Labradar increases with distance. This is reflected in the speed measurements, by averaging several shots we should have an acceptable speed measurement, assuming that the error responds to a Gaussian dispersion.
Bob, assuming we could get reliable data, how would you propose calculating the drag coefficient or a ballistic coefficient for a drag curve model.
-
Without a doubt we cannot have speeds that increase with distance, we must consider that every measurement has an error and that error in the case of the Labradar increases with distance. This is reflected in the speed measurements, by averaging several shots we should have an acceptable speed measurement, assuming that the error responds to a Gaussian dispersion.
Bob, assuming we could get reliable data, how would you propose calculating the drag coefficient or a ballistic coefficient for a drag curve model.
If you want to obtain a drag curve, there is a fixed method which you should use.
1. obtain your data on a day when there is very little or no wind. Wind can have a surprising effect on your results. Ensure you either measure or obtain the meteorological conditions at the time of your test to enable you to calculate the air density.
2. Get data on at least 10 shots.
3. Take the raw data from the LabRadar csv files. Do not put a curve fit through the velocity range data. I say this because the shape of the drag curve you calculate will be first order dependent on the shape of the curve you decide to fit to the velocity range data, and will not be representative of the true drag curve shape. This is something I found out after analysing literally thousands of measured velocity range data files for bullets.
4. Using the simple flat fire equations to solve for drag coefficient (as long as you are testing over a level distance, the corrections for gun and trajectory elevation are not worth the extra trouble) analyse all of the data, moving along through the trajectory for each time step using the air density at the time and place of your tests. Do not worry at this point about negative drag coefficients or the spread of results. Obtain lists of drag coefficient vs bullet speed for each of your measured shots. I always used time and velocity as the basic data for analysis, as these are the basic, measured characteristics produced by the radar.
5. Combine all the data calculated above into one sorted list of drag coefficient vs speed or Mach number. Using Mach number is more accurate than speed.
6. Now is the time to use either a curve fit in your software, or draw a mean line through your data by eye, which can be better as you can see and ignore data which is obviously in error.
The above sound laborious, but experience has shown it is the best method, and once you have the software and data sheets set up, it does not take that long.
Good Luck.
-
ballisticboy thanks for the methodology, I can do these calculations. But I need to tell me which equation "simple flat fire equations to solve for drag coefficient" you are referring to.
The literature that I have read does not provide a ballistic model of equations to use for the solution of gunshots.
Could you guide me where to look for that information or provide it to me?