Chronograph Madness

Jump to latest Follow Reply
G

Gwhite

Member
Jun 26, 2019
13
2
MA, United States
I am the team "armorer" for a US college pistol team, and I'm learning how to repair competition 10 meter air pistols. An important part of the process is measuring & adjusting velocities. In the past, I have used Combros, which are spec'd to be 1% accuracy. I have two at home, and we have one at the college. I haven't managed to shoot one yet, but it would certainly be easy to do. I recently picked up one of the Chinese HT-X3006 units, which claims 0.5% accuracy. I've 3D printed a bunch of adapters so I can easily pop it onto a variety of pistols.

https://www.thingiverse.com/thing:6977381

It's easier to use than the Combros, and the mounts I designed make it pretty idiot proof in terms of not shooting the sensors.

I tested the HT-X3006 on a Benelli Kite I had just rebuilt. If they are up to factory spec, the velocity shouldn't vary by more than +/- 1 meter/sec. I was seeing variations of up to twice that. I didn't know if it was the pistol, or the Chinese chrono. I tested the pistol with all three Combros, firing 25 shot strings with each. Nothing matched up very well:

ChronographAve. VelocityStandard Dev.
HT-X3006
138.04 m/sec​
1.668 m/sec​
Combro #1
133.60 m/sec​
1.433 m/sec​
Combro #2
134.30 m/sec​
1.300 m/sec​
School Combro
135.10 m/sec​
1.758 m/sec​

The Combros are roughly within 1% of each other, but the "more accurate" HT-X3006 is a couple percent above ALL of the Combros. The standard deviation numbers are all big enough that I think the pistol has a problem. The catch is that I have no idea which chronograph to use to test it when I try to fix it.

I also own a LabRadar V1 Doppler radar chronograph, which is supposed to be good to 0.1%. The little chronos are WAY easier to use, so I figured I could set up the LabRadar and use it to come up with calibration/correction formulas for the others. I could shot through the small chronos, and use the radar to measure each shot at the same time, which should give me really good data. Or so I thought...

I set up the LabRadar in my basement shop, and spent several hours trying to get ONE valid reading. In order to get 10 meters, I have to shoot though a doorway, and the place is very cluttered. After playing with all sorts of settings, I concluded that I was getting too many reflections off of various surfaces, and/or the distance was too short.

Today, I dragged the LabRadar and the HT-X3006 into the school 50 foot range, which is relatively open. I could immediately get velocity readings, and I tried to get some comparison data between the two. I need to plot all the data, but here are a few samples (all in fps):

TestLabRadarHT-X3006
1463480.8
2389444.1
3417443.1
4406461.1

The LabRadar varied all the way from 389 to 477 fps (an 88 fps spread). The HT-X3006 was considerably more consistent, going from 441 to 480.8 fps (39.8 fps spread). The "match" between velocities on the SAME shot was universally horrible.

I understand the "physics" of the HT-X3006, and it's pretty basic. There's only so much that can go wrong. I also used to work as a microwave engineer, and Doppler radar is pretty well established technology, but picking up a tiny pellet seems to be a problem. I now have FIVE chronographs, and I can't trust ANY of them. At this point, the $25 Chinese box is giving me much more believable data than the filthy expensive 0.1% accuracy radar.

Given that they are cheap, I may just get several of the HT-X3006's and see how they compare. I can get four for what I paid for one Combro. I don't actually need to know the absolute velocity super accurately, but I need to be able to measure the consistency to at least 1 fps or better.

I took a look at the Garmin Xero C1, and it claims slightly better than 1% accuracy, but there's no info on how repeatable they are. I also have no idea it it would work in my shop at home. The convenience of the Combros and the HT-X3006 is hard to beat. I've measured a number of air pistols that fire consistently around +/- 1 fps with Combros in the past, so they appear to be pretty repeatable, even if the absolute velocity is slightly suspect.

I could get a larger conventional chronograph, but I really don't have a good place to set one up at home. I even have an old Oehler 35P with three sensors, but that's WAY to big to set up in my shop.

So, now that you've suffered through my tales of woe, does anyone have suggestions for a chrono I can actually trust? I'd blow the big bucks on a Garmin if I KNEW it would work well with pellets, especially if I could use it at home. Based on my experience with the LabRadar, I don't trust a Doppler system in that environment, which leaves me with optical time & distance gadgets.

Thanks!
 
If you haven't already, download the labradar individual shot traces from the sd card. With the labradar set for subsonic range, the trace should be every 0.002 seconds. It will give you time, distance, velocity, and most important Signal to Noise Ratio (SNR). If the first few yards it picks up are not absolutely consistent (both velocity drop every .002 seconds and a nice clean line of SNR), you probably have too much interference in your indoor range to use it there. If you had your unit set for high power, go to the settings and set it for low power and try again. I get better data at one of my ranges on lower power due to all the interference on that one, my data there is only good to a max of 15 ish yards on a .22 size projectile, and not much past on .429 projectiles. Really bad set up for radar there but works for my load development just fine. It only works well for me there because I know where the center of my beam is, and I can accurately aim my labradar to less than 1.5 feet off center at 100 yards, literally on my bad range if my aim were off 2.5 feet at that distance I would not get any good data, and lots of no tracks. Interference can be a real b*tch, and infinitely worse if you are not shooting through center of beam on labradar. FYI, I love my Labradar and wish I hadn't waited to buy it for a couple years after it came out. It works all the time, the same every time, takes me less than 2 minutes to walk outside and set it up with near perfect aiming. Infinitely better than any optical chrono I've used going back to the 1970's.
 
Upvote 0
Thanks for the feed back! I didn't bother with an SD card. I'm new to the LabRadar, and didn't realize it stored SNR data on the card.

I plotted the results:


4-15-25 LabRadar vs HT-X3006.jpg


The first shot looked very promising, and then it rapidly went to hell. I started out at the low power setting, and had very reliable acquisitions. At the end, I switched to the high power setting. It failed to acquire about half the time, and I only collected 3 shots of data before I gave up. The accuracy of the results appears to be much worse. One issue is that the ceiling of our range has angled steel baffles, and the vertical beam width of the radar is wide enough that I may be getting reflections off those. However, they are stationary, and shouldn't interfere with receiving the Doppler shifted signal.

I live in New England, and if I can't use the LabRadar in a relatively open indoor 50 foot range, it's pretty much useless during most of the school year, i.e. winter.

I also checked the specs on the FX radars, which are at least specifically designed for air guns. They claims +/-2%, which is not very useful for tuning purposes compared to the Combros or HT-X3006. If I knew if would be easy to use (indoors) and consistent, it might still work.

I'm going to send my plot to LabRadar. I'll see what they recommend, and probably try it again with an SD card.
 
Upvote 0
Last outing my friend, having partially set aside his rifle with chrono ( FX V2 ) attached.

So something must be flying around over our range at greatly alternating speeds, CUZ suddenly it started to bark out speeds, from in the blowpipe category and up, but the HIGH one was like 32850 fps

Ill have him MSG the screenshot to add.


SAVE_20250416_151424.jpg


MY FX V1 work fine in the living room shooting 9 feet to the pellet trap
 
Last edited:
Upvote 0
What might be causing some problems is vapor trace as far as the barrel mounted units are concerned. Now before you tell me your air is dry you still have ambient air in the barrel and depending temps and humidity it could be your problem. If you really want to see a barrel mount unit go nutz try a co2 unit. Back in the day I could see that vapor trail on some indoor ranges as well as outside at times, and yep the chrono would give false readings ( not barrel mounted) a far as radar units- i am always reminded of the 100,mph palm tree defense clocked on a police radar gun in some traffic violation case years ago.
Radar is a bounce back signal so cluttered area can reflect, in extreme cases in an inclosed area stratification of temperatures in a range can also cause problems. Operative word on that is hash. Brand new Orings and such take awhile to settle in and a spic and span clean barrel will typically dance around a bit as well.
just some musing from an old master in many disciplines.
 
Upvote 0
I'm using moderately dry air from a SCUBA tank. When I have a properly tuned up pistol working really well, I get +/- 1 fps readings from a Combro. I need to go through my notes and see which pistols could do that, and use them for my testing. I've repaired about 20 pistols in the last year or two. About 50/50 Benelli Kites and Hammerli 480K's that were converted to use Hammerli AP40 cylinders. The Hammerlis often have REALLY low velocity variations. Better than the Steyrs I've tested.
 
Upvote 0
I recently purchased the Garmin but have not tested it yet. I plan on using it for .177, .22, .25 and .30 pellets airgunwise. I have a very finicky V1 FX chronograph and it really dislikes .177 pellets. I am hoping the Garmin does a better job. I have watched a lot of reviews on the Garmin and no one mentioned missed shots. Fingers crossed.
 
Upvote 0
I’m shocked to read about any success with the Labradar and airguns. I was under the impression that getting such small calibers to register required a microphone or trigger attachment to even have a chance. I can’t comment on the accuracy claims or actual performance of the other chronographs you’ve tried, but will say the Xero has exceeded my expectations, captures almost everything, and operates in conditions that the Labradar can only dream of.

I set up my own test, indoors, 15yds, normal basement ceilings and assorted clutter, to compare a Pro Chono optical chronograph with light kit, a Magnetospeed, and the Xero. I could not get the Magnetospeed to reliably pick up .177 pellets, however I was able to compare .22, .25, and .30 cal across all three units. I believe for a total of 50 shots across the three calibers and different power levels, the single largest discrepancy was 6fps. Typical variation was 2-3fps across all three chronographs. It was actually pretty impressive.

I can understand wanting a system that instills confidence in your process and results, and eliminates the gun’s performance as a variable. That said, how much difference does 3fps make at 10m? How much difference does 6fps make? I would think that the Xero would give you satisfactory results. Maybe the new Athlon release will be worth a look as well.
 
Upvote 0
Thanks for that info! Given my results with the LabRadar, I'm very suspicious of radar chronometers. At least it sounds like the problem with short ranges & reflections isn't an issue with the Xero. It would be interesting to see side by side results of a Xero and an FX (or an Athlon) watching the same pellets being fired, one on each side.

I am using the air gun trigger they sell for the LabRadar. It appears to trigger reliably with that. The problem is that the radar isn't seeing the pellet well enough to get a good reading. In the 50 foot indoor range, if I set it for full power, I'd get an "acquisition" failure about half the time. In my (very cluttered) 10 meter basement, I never got a single valid reading even at the low power setting.

One issue may also be the lower velocity of air pistols. That means the Doppler frequency shift is smaller, and that could make the receiver less sensitive due to noise issues.

For a 525 fps 0.177 caliber pistol with 7 grain pellets, a +/- 3 fps velocity difference at 10 meters will produce a ~ 0.5 mm vertical spread. According to Benelli, +/- 1 m/sec is considered the upper acceptable limit for their Kite pistols. The test targets that come with high end Olympic grade pistols (Walther, Pardini, Steyr or Morini) typically show results better than that. That level of variation would probably be considered unacceptable for an Olympic grade air rifle, where the 10 "ring" is actually a dot 0.5 mm in diameter.
 
Last edited:
  • Like
Reactions: Tumblinginflight and beerthief
Upvote 0
From your list, the only chronograph I have experience with is the LabRadar, and that is only about six months worth. The biggest thing I can say is that I absolutely agree with @karl_h - you have to at least look at least a few traces of detailed data for pretty much every shooting situation (location and projectile used) to understand the robustness of the data that the LabRadar is giving you. It is important to realize that it will calculate speeds for you from the data it has, but those calculations may be meaningless if the S/N is not good enough. It is even possible for pellets to get faster down range within the trace data - which clearly tells you that we are dealing with garbage data. Note that I have not seen that yet reported out on the main screen for the calculated values reported there (and those are calculated values) but that does not mean that the data under them are not sketchy. You'll only know by looking at several traces and seeing how far you can trust the data. Personally, I use a S/N level of about 25 as the cutoff for trustworthiness, and I won't use data from ranges beyond where I start to see the S/N drop below that level from a at least a few individual strings.

Second, the S/N is almost always worse indoors than outdoors (for a given range) as there tends to be more reflective surfaces around. While it is advantageous that the reflective surfaces are not moving indoors (unlike grasses or branches moving in the breeze outdoors) the amount of reflections creates noise in and of itself. I know it's not something we can easily do, but if you want the best data, shoot outdoors in an open area with the unit angles upward so the ~7 degree conical beam hits nothing but air (so about a 10 degree upward angle). Of course that adds a bit of gravitationsl resitiance into the equation, but the data will be very clean . . .

Personally, I've not come up with a good way to calculate the muzzle velocity that LabRadar spits out as a value - my attempts to calculate it from the raw data always yield a difference of a few fps - not that matters much in the big scheme of things. But knowing that the data being used inthe trace is robust is the most important thing.

As for the Chinese unit you have, the biggest thing that would concern me about it is the possible "language" issue on the claims. Honestly, I would expect the 0.5% "accuracy" claim to be more about reapeatability than accuracy. But it could be something completely different, like the mathematical varience expected from the reported clock cycle time and distance between sensors - I know that was the value that was reported on the common foldable "Shooting Chrony" models.

Sicne you are working with competition pistols, I would expect that the good ones are very stable in speed. You could always take the "best" one of those that you have available and do a bit of a "reverse test" using it. For that, you could take weight sorted pellets (to minimize speed variation) and test for which unit gives you the most repeatable results from that pistol, in terms of SD of output. It would tell you nothing much aabout accuracy, but might tell you something on repeatability.

Good luck!
 
Last edited:
Upvote 0
Thanks for the very detailed info! It would be nice to know the velocity, but the range for the Benelli Kites is 135-145 m/sec, so it's not super critical. What's important is the repeatability. Given that I've measured +/- 1 FPS on some air pistols with a Combro, they have to be pretty good, and I wouldn't expect the HT-X3006 to be worse, but who knows?

I repaired a bunch of Hammerli air pistols last year, and I have at least a few that were +/- 1 FPS, but I was only testing over 5 shots strings. At least with the Kites, I've found that they tend to vary a lot more than that after a major rebuild, but the good ones will settle down after about 15 shots. Here's a before & after on one where I rebuilt the firing valve system:

1-2-25 SN 00547R Before & After.jpg


This data was taken with a Combro, and I've never seen any reason to doubt their repeatability. However, there have been multiple reports of Combros reading lower velocities compared to other chronos. Apparently they can also drift downward if the batteries aren't fresh. That's why I'm leaning towards the Chinese one for most of my testing (besides the ease of use). I agree that the +/- 0.5% accuracy might be based on clock crystal specs, etc. For a small device, the spacing of the sensors is probably a bigger issue. The Combros were calibrated at the factory (against what?), and there is an adjustment parameter you can set. However, it only has a +/- 1% range. I have no idea if the Chinese ones have a calibration parameter buried inside, of it they rely on precisely locating the sensors. In both the Combro and the HT-X3006, the sensors are only about 60 mm apart, so they have to be soldered in place within ~ 1/2 mm to give close to 1% accuracy. That's not too hard to do, so +/- 0.5% the HT-X3006 claims (and the Combros claim to meet so they can guarantee 1%) is not unreasonable.

I thought the smaller radar Chronos might use a higher frequency than the LabRadar, which operates at 24 GHz. There are automotive radars that run at ~ 75 GHz. Apparently the small ones still run at 24 GHz, which means their beam is MUCH wider than the LabRadar. That would explain why they are much more forgiving in terms of alignment & acquisition, but the energy is going to be spread out a LOT. What I can't figure out is how they keep the SNR up with that much less energy being reflected back. One difference is that the small ones only worry about muzzle velocity, where the LabRadar is designed to track the bullet out at a fair distance.

It looks like the Athlon is just shipping now, so there's not much real user info on them, especially for air guns.
 
Upvote 0
@AlanMcD SNR is more complicated than just a value. For instance, you mentioned a SNR below 25 in your situation can easily lead to be bad data. At my good range at home, I need near an order of magnitude better than that (27), and on my bad range at home I need a SNR of 30-32+ to get good data. Interference is a bitch. Fyi, a properly aimed labradar will give you over 40 SNR when first picked up on .22 projectiles, both 22lr and 22 pellets. Kind of tough to do if you haven't mapped your unit with an extremely accurate way to aim it.
 
  • Like
Reactions: AlanMcD
Upvote 0
One note on the LabRadar beam. The specs say the beam is 7.6 degrees by 18.5 degrees, but they don't say which is vertical & which is horizontal. I asked the factory, and the tech support person had to double check. I had assumed the 18.5 degrees would be horizontal to help with acquisition. WRONG! That the vertical spread, apparently to track a bullet at ranges where there is significant drop.

I've wondered if I rotate the unit 90 degrees if I could get better results at short range...
 
Upvote 0
@Gwhite, cut those degrees by about 1/10th and you are in the sweet spot, and it makes a massive difference. Literally, a 2.5 foot error in aiming the center of the beam at a target 100 yards away drops the radar return by several orders of magnitude. If you have never seen your SNR hit 40 or above on a .22 when picked up, you have never been accurately aimed/and/or misspostioned shooting position. On mine, and after looking at construction I'm sure there is a wide range on every unit out there, the center of the beam is not perpendicular to the face of the unit. Your idea of rotating the unit is unfortunately pointless, the radar beam is very weak at the edges of their specs. You really need to map your beam and have a perfectly repeatable way to aim, a pen in the notch is not repeatable, not even close. Not to mention the software expects the projectile to enter from the side, don't know what effect that would have, probably not good.

I also have seen several people using theirs and they are not really close to the offset to the side they put in the settings, you really want to be within a couple inches of what you enter on the offset in the settings. A friend, more of an aquaintance that we get along well, lets me play with his expensive toys once or twice a year. Expensive as in many would be a decent yearly salary for many people. He had problems with his labradar, I got him straightened out on a long weekend at his place. All his problems were his usage, not the labradar. He was ready to use it as a target on his 1800+ yard private range.
 
Upvote 0
I have a 3D printed sight with a tube on top, but if the LabRadar beam is off center from the face, that doesn't help much. When I get time, I will drag mine back to the 50 foot indoor range and try again with an SD card so I can monitor the SNR. I really need to fix more pistols, not become an expert on the idiosyncrasies of my chronographs...

I've ordered another HT-X3006, but with the Wi-Fi option so I can collect long strings of data and get them into a spreadsheet more easily. I may try to put the two in series to see if they read the same on a given shot. That would give me more confidence that they are working well. It will take a few weeks to get here from China, but that gives me time to 3D print a bracket to hold a pair. I will test them with one of the Hammerli pistols that showed very small velocity variations.

Given that I apparently can't use the LabRadar in my basement shop, it has VERY limited utility for me. Between the Garmin, the FX, the Athlon and the LabRadar LX, there are quite a few options for small radars with wider beams. I have yet to hear anyone bad mouth the Garmin, and it sounds like a few people have had success with one at short range indoors. The LX and Athlon are both so new that there's very little real info, especially for my use case.
 
Upvote 0
I'm not sure that is much of an issue with the air pistols we shoot. The mounts I have 3D printed for the Combro put the muzzle about 2 cm back from the first sensor, and the mounts for the HT-X3006 are similar. Because the turbulence of the air exiting the barrel can affect the accuracy, most competition pistols have an "air stripper" system that vents the gas away from the path of the pellet. In many instances, this starts well back from the actual muzzle.

HT-X3006 Morini Mount (Side).jpg


I've been careful to make sure the pistol mounts I've printed don't block the venting system on the pistols (give or take the occasional rubber band...).
 
Upvote 0
karl_h said:
@AlanMcD SNR is more complicated than just a value. For instance, you mentioned a SNR below 25 in your situation can easily lead to be bad data. At my good range at home, I need near an order of magnitude better than that (27), and on my bad range at home I need a SNR of 30-32+ to get good data. Interference is a bitch. Fyi, a properly aimed labradar will give you over 40 SNR when first picked up on .22 projectiles, both 22lr and 22 pellets. Kind of tough to do if you haven't mapped your unit with an extremely accurate way to aim it.
Click to expand...
Thanks for the feedback. I'm aware of the complexities of SNR, but I've only used my LarRadar in one location so far, other than some tests indoors. I'll be sure to check out how it can change when I get the chance to use it in other locations, and I'll do some work on mapping out the best alignment of the beam . . .
 
Upvote 0
You must log in or register to reply here.
Top Bottom
Back
Home
Forums
Resources
Menu