Calculating BC with LabRadar. It works!

[entoptics]

Picked up a LabRadar a couple days ago. It's a very nice machine.

Quick Review (though you've heard it all before).

Pros - Easy to setup, appears very accurate (tested against ProChrono), can run two rifles side by side with a buddy without repositioning, amazing data collection, so far no problem reading 300 WM and 5.56 out to about 80 yds (sometimes 90-100).

Cons - Stupidly expensive, needs a remote (wiggles when you press buttons, so you have to re-aim every string, and trying to control it from prone is just silly), no bluetooth to smartphone (yet), doesn't include MacDonalds straw for aiming, can't revisit a string to add more shots, and did I mention it's stupidly expensive?

Anyway. To the point in the title. You can absolutely calculate B.C. using the data it collects. And surprisingly accurately. Unfortunately, it's not as trivial as setting your distance points and plugging and chugging in JBM when you get home.

The machine records it's acquisitions in a CSV file. It appears to read every millisecond as the round travels downrange. It outputs a range, velocity, and signal:noise. I took that file and graphed stuff out, and noticed that sometimes there were obvious fliers, and minor scatter overall, in the data. This got me thinking.

Instead of taking the raw machine readings at two distances and plugging them into JBM, I fit a line to the data using Excel. I calculated theoretical "Near" and "Far" velocities and distances, using the fit line equation, and popped them into JBM. Suddenly out popped BCs that were scarily close to the manufacturer's advertised (Hornady 208 ELD-M and 212 ELD-X).

I only experimented with 2 shot strings, using atmospheric data from NOAA for a station 10 miles away, and my calculated B.C.s were ≤ 5% from what Hornady quotes, and for a couple shots, within 0.5%. Both G1 and G7 seemed to be darn close.

In summary, I'm thinking that with better atmospheric data (e.g. recorded on site with a kestrel/smartphone/weather meter/etc), and a half dozen shots for good statistics, one could get pretty precise BC results with the LabRadar.

I'll keep up the experiments, and potentially make an Excel template to distribute if anyone is interested.

 

[Pdvdh]

I've also back calculated bullet BC values. The more bullets fired the more confidence I have in the calculated BC values.
I've also observed the "noise" in the CSV files. So I have more confidence using the velocities from some additional shots fired @ 0-20-40-60-80yds. I don't mess with the CSV files anymore. I presume the LabRadar engineers have already applied a good mathematical fit from the CSV data which helps focus thru the background noise and yields best fit velocity values at the pre-programmed distances.
We all know there is some decay (lowering) of BC value as bullets slow down in flight over long distances. So the BC values I've calculated are considered good values for closer ranges, like out to 400yds. I have not tested them by collecting down range velocity at 800yds-1000yds. Nor have I drop tested the LR based BC values by measuring drops on target at Long yardage. For those wondering how I collect bullet velocity @ 1000yds - suffice to say it is not with the LR. I use a triplicate optical chronograph setup.

I find the LR particularly valuable for generating BC values for custom bullets which have not been BC verified by the manufacturer with Doppler radar or some equivalent velocity measuring equipment.

 

[Pdvdh]

Cons - needs a remote (wiggles when you press buttons, so you have to re-aim every string, and trying to control it from prone is just silly), no Bluetooth to
smartphone yet...

This is a significant negative to the LabRadar. When I purchased mine in November/December 2017, I was told the Bluetooth App would be made available soon. Well they've been propagating this fairytale for more than one year now. My advice is don't buy one until the remote arming of the device is available. I've about concluded there must be a serious problem with implementing the bluetooth remote arming feature. And it may never happen. I have no inside connection. But really, how long does it take to create a program to arm via Bluetooth?

 

[entoptics]

...I presume the LabRadar engineers have already applied a good mathematical fit from the CSV data which helps focus thru the background noise and yields best fit velocity values at the pre-programmed distances...

...I find the LR particularly valuable for generating BC values for custom bullets which have not been BC verified by the manufacturer with Doppler radar or some equivalent velocity measuring equipment.

I assumed as much too, and strangely, for some of the downrange calculated velocities (i.e. at the distances you set in the software), this is the case, but the V0 velocity is often way off the line for some reason. Of the half dozen shots I calculated, most were over 5-10 fps different from the line projection. In most cases though, the V3/V4 (sometimes even V5) output from the Labrador was within tenths of an FPS from my linear regression calcs.

I've attached some graphs that show what I'm talking about. Blue Squares are CSV data, Red Triangles are LabRadar V0-V5 output, and Green square is my calculated V0 velocity using my own regression.

It's frankly a bit weird, and perhaps concerning, that the algorithm LabRadar is using isn't internally consistent. Especially considering the linear regression calcs are closer to the expected velocities based on the load data, as well as the more accurate BC calcs returned vs the LabRadar calculated data.

In the case of a 4 shot string for a 208 ELDM, my calculated velocity was 7 fps slower and had an SD that was 3 FPS lower than what LabRadar calculated for the same string.

 

[Pdvdh]

How do you think they get the correct MVs? Most all comparison of MV data I see posted when the LabRadar is run concurrent with a MagnetoSpeed or the Oehler 35P indicates its V0 velocity is accurate.

Maybe the algorithm they employ knows more about best fit of the CSV data to an accurate MV than we do?

When I calculate bullet BC value from the LabRadar velocities, I'll average the bullet velocities recorded at each programmed distance. Typically V0, V20, V40, V60, and V80yds. If I record 3 bullet velocities, I'll use the average of the 3 velocities at each distance and then model those velocities for BC value using Coldbore 1.0 ballistic software. This helps minimize the error due to any one recorded bullet velocity. I even do this when the loads are at different powder charges, producing differing MVs, during load development. Not a problem because BC is determined from the change in velocity over distance. The differing MVs don't impair the accuracy of the modeled BC value.

 

[TwoMore]

This is a significant negative to the LabRadar. When I purchased mine in November/December 2017, I was told the Bluetooth App would be made available soon. Well they've been propagating this fairytale for more than one year now. My advice is don't buy one until the remote arming of the device is available. I've about concluded there must be a serious problem with implementing the bluetooth remote arming feature. And it may never happen. I have no inside connection. But really, how long does it take to create a program to arm via Bluetooth?

thanx, that is good to know, I am getting ready to buy the labradar I may try to hold off a bit

 

[entoptics]

How do you think they get the correct MVs? Most all comparison of MV data I see posted when the LabRadar is run concurrent with a MagnetoSpeed or the Oehler 35P indicates its V0 velocity is accurate.

Maybe the algorithm they employ knows more about best fit of the CSV data to an accurate MV than we do?

I too have heard that they are very close to other quality chrony data. They claim ± 0.1% accuracy, and the variance I am seeing between a linear fit and their calculated numbers is no more than 0.5%, which is probably as accurate as any 35P or Magnetospeed.

If you look at the graphs I posted though, it's quite clear what's influencing the variation from the linear regression. Their algorithm must be weighted towards readings near the set distances. Perhaps some sort of rolling regression of the points within a few milliseconds of the programed distance.

Note V5 on graph 1 is below the line due to the cluster of spurious readings at the end of the flight tracking.

Note V5 on graph 2 is below the line due to the spurious reading returned at ~75 yds.

Note V1 on graph 3 is way off the line due to the spurious reading returned at ~10 yds.

Note V1 on graph 3 is off the line high due to the spurious reading at ~18 yds.

Note that when the tracking is good (generally between ~30 yds and ~60 yds) The V3/V4 values lie right on the regression.

I understand that external ballistics is quite complicated, but based on my results (a small sample to be sure), I think the internal calculations may be subject to more influence from instrumentation "noise" than they should be.

They had to pick an algorithm that would be good for all types of speeds and bullets, and perhaps it's a worthwhile compromise, but might result in some small error in specific conditions.

I'll run some more data through and see if obvious patterns emerge. Regardless, the unit is still the best chrony available for the enthusiast in my opinion.

 

[Pdvdh]

You're conducting some good, and interesting, analyses.

I'm sure there comes a time for compromise in the manufacture of these units, involving R&D time, parts & production costs, and profit.

We know there are better Doppler radar units available. But we can't afford them, or maybe even transport them. My primary "want" for the LabRadar is the Bluetooth arming App, so I can 'arm' the unit when shooting from prone position.

 

[DocUSMCRetired]

They had the Bluetooth app and up and running at NRAAM this past weekend. More detailed than I expected it to be, and it was fully functional.

 

[Pdvdh]

That's good news for LabRadar owners. Did they announce a release date?
Was this an iPhone or android -based App?

 

[DocUSMCRetired]

That's good news for LabRadar owners. Did they announce a release date?
Was this an iPhone or android -based App?

Both.

I think they are trying for before the KO2M.

 

[Bigdad]

I got an email today from LabRadar stating the app is available from the Google App Store and firmware is ready for download from their website for Android only. I'm going to give it a go tonight when I get home. Hoping for a good outcome.

 

[Pdvdh]

Yup. Wish my phone was Android.

Question? Could a guy buy a used Android OS smart phone with Bluetooth, and use it solely with the Labradar without buying another wireless cell phone plan?

Any techies know if that should work?

 

[Bigdad]

You can buy a $50 android tablet at Walmart, should work fine as long as it has Bluetooth. I was thinking about doing this just to get a larger screen.
https://www.walmart.com/ip/RCA-Voyager-7-16GB-Tablet-Android-6-0-Marshmallow/53990885