Home Made Bearpaw Spine Tester

dvd8n

Supporter
Supporter
AIUK Saviour
It has given me a weird issue though; previously the display worked flawlessly, but now it intermittently won't show line 2. It'll work, then suddenly stop working for a while then mysteriously start again. It shows all the hallmarks of being a hardware issue, except that the demo display programs that come with the display always work without fail. Which makes me think that it has to be something really insidious in my software.
Ok, I think I've got that sorted out. Now to try to cut some rectangular holes in an aluminium box that don't look like they were cut with a cold chisel by a gorilla.
 

Rik

Supporter
Supporter
Finding ways to complicate/simplify:
If you have a shaft with a "known spine" (one you trust) then you could calibrate from that instead of a reference weight? Make everything relative to that shaft.

Though speaking of relativeness, if the aim is to compare shafts, then no calculation is needed, as the figures off the scale will give relative stiffness immediately. It's just a question of knowing how big a difference (in weight) is important.

For accuracy, if kitchen scales are a little coarse, would grain scales be too fine?
 

geoffretired

Supporter
Supporter
Some great stuff going on there... Wow!
About the weight used.... if you have a set of arrows, and they shoot well from your bow, Then you could assume they are " a right spine match" for your bow.
If you used a weight that wasn't exactly what Easton uses .... does that matter? I would guess that if all the spines matched each other, that would show in you results even if the "number " wasn't what Easton would get with their testing. An odd shaft would still show up weaker or stiffer than you want.
 

dvd8n

Supporter
Supporter
AIUK Saviour
Finding ways to complicate/simplify:
If you have a shaft with a "known spine" (one you trust) then you could calibrate from that instead of a reference weight? Make everything relative to that shaft.
I did have a quick go at putting in a compensation factor calculation in as part of the calibration routine, basically a multiplication factor based on a known good 500 arrow.

It made things worse for all the other arrows 🙄

I'm not sure why, it could have been a bug in the software or it could have been that my known good arrow wasn't as good as I thought. The latter is almost certainly true as all my 500 arrows have been shot thousands of times. Maybe next time I need arrows I'll add a quality unused full length 500 arrow to my order to use as a reference.

Anyway, I decided not to take it further as it was turning into a time sink. I can always go back to it later if there's a third lockdown (or are we up to three already?)

Though speaking of relativeness, if the aim is to compare shafts, then no calculation is needed, as the figures off the scale will give relative stiffness immediately. It's just a question of knowing how big a difference (in weight) is important.
You are absolutely right.

For accuracy, if kitchen scales are a little coarse, would grain scales be too fine?
Not so much too fine as too delicate. You should have heard the crunching noise from mine when I tried it 😢
 
  • Like
Reactions: Rik

dvd8n

Supporter
Supporter
AIUK Saviour
Some great stuff going on there... Wow!
About the weight used.... if you have a set of arrows, and they shoot well from your bow, Then you could assume they are " a right spine match" for your bow.
If you used a weight that wasn't exactly what Easton uses .... does that matter? I would guess that if all the spines matched each other, that would show in you results even if the "number " wasn't what Easton would get with their testing. An odd shaft would still show up weaker or stiffer than you want.
You are totally correct. Easton's (sorry, ASTM's, hem-hem) spine numbers are really just an arbitrary number which is being used as a proxy for static spine which is itself a proxy for dynamic spine. But it's pretty much the industry standard.

But if you're just wanting a number for comparison then you could do what I did on the kitchen table with the kitchen scales in the second post. It'd be fine. I guess that you could even build the scales into a frame with an arrow support glued to it as a low tech solution if you're happy with a result in grams. Or, slightly less low tech, gut the scales for parts and make it look a little less janky.
 
Last edited:

dvd8n

Supporter
Supporter
AIUK Saviour
Actually, having a 'house standard' for spine would be quite liberating as it would mean that you wouldn't have to set the supports at 28" or the pillar bang on 1/2".

I was surprised at how convenient the alternative ASTM spacing of 23" was - I found that I could test fletched arrows, arrows with wraps and pretty much all shorty arrows. I didn't use 23" in the final build as I wasn't getting readings with great accuracy - they were repeatable but not not that close to the ASTM spines. But that was admittedly before I realised how critical all the positioning was.
 
Last edited:

Rik

Supporter
Supporter
Actually, having a 'house standard' for spine would be quite liberating as it would mean that you wouldn't have to set the supports at 28" or the pillar bang on 1/2".

I was surprised at how convenient the alternative ASTM spacing of 23" was - I could test fletched arrows, arrows with wraps and pretty much all shorty arrows. I didn't use 23" in the final build as I wasn't getting readings with great accuracy - they were repeatable but not not that close to the ASTM spines. But that was admittedly before I realised how critical all the positioning was.
Yes... I would not be able to use the ATA 28" gap for any of my shafts. 26" is tight, as I generally cut to 26.5...
 

dvd8n

Supporter
Supporter
AIUK Saviour
I worked out the design of the circuit to construct on Vero board:

IMG_20210408_174201.jpg

I then marked up the board to place the components. If you look really hard, you might be able to spot the 'deliberate' mistake, and it's not that I had to squeeze up some components to allow for the notch for the battery box.

If you can't be bothered, it's that the connectors for the battery and power switch are misplaced by one row. Never mind; I caught it in time.
 

dvd8n

Supporter
Supporter
AIUK Saviour
And with the box closed up.

IMG_20210409_204906.jpg

I'm still waiting for some connectors to connect the control box to the load cell, then I can put the end plates on the box and I'm done.
 
Last edited:

dvd8n

Supporter
Supporter
AIUK Saviour
????????????????????????????!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

IMG_20210409_205016.jpg

Arrow spinerow; ???? Really?

I guess the software still needs a little work 🤦‍♂️
 
D

Deleted member 7654

Guest
Great project :)
Allows me to get some vicarious electronics and mechanical pleasure (without the work :ROFLMAO:)
Del
 

dvd8n

Supporter
Supporter
AIUK Saviour
Ok, fixed the bugs and bolted it together. I'm still waiting for DIN connectors to allow for easier storage but I just hard wired it - I was sick of it lying around half assembled making the place look untidy.

IMG_20210412_165706.jpg

IMG_20210412_165830.jpg

I'll revisit it when (if) the connectors ever arrive.

Anyway - time for a post-mortem.
 

dvd8n

Supporter
Supporter
AIUK Saviour
#1 - Buy the Bearpaw

This project didn’t turn out cheaper than the Bearpaw. None of the items that I used were expensive, but costs mount up. A lot of this was down to my design decisions. I could definitely make a cheaper Mk II but then I really don’t need a wide selection of home-made arrow spine testers.
 

dvd8n

Supporter
Supporter
AIUK Saviour
#2 - Buy the Bearpaw

The Bearpaw has facilities for weighing arrow heads. The scale that I made struggles to get within 1 gram of accuracy. This is fine for spines, but is hopeless for weighing an arrow head. Now, I used a 5kg strain gauge. I could probably swap it out for a 1kg strain gauge which would have less range but more precision – it would be marginal for stiffer arrows but feasible. However that would only get me 5x the accuracy, which still wouldn’t get me down to a grain. Bearpaw must be using better strain gauges than I have access to / can afford.
 

dvd8n

Supporter
Supporter
AIUK Saviour
#3 - Buy the Bearpaw

The Bearpaw has a strain gauge at both ends. I only used one at one end, and assumed that the forces on each column would be the same. It turns out that that assumption only holds if you are disciplined about pressing the arrow down dead centre. I had to slim down the size of the central pillar as much as was feasible to help with pressing it centrally and it can still be a bit finicky. Two strain gauges would do a lot to mitigate that – instead of guessing the force on the second column and getting it wrong, I could just read it. Two gauges would be much easier to use at a design level and at a user level.
 

dvd8n

Supporter
Supporter
AIUK Saviour
#4 - The End Columns

My end columns are hugely over-engineered. I made the columns really sturdy and put quality bearings in them as that’s what the ASTM specifications said to do. But a ballpark calculation using triangles reveals that the arrow is going to move less than ten thou at each end. Really, those bearings aren’t moving any appreciable amount at all. Bearpaw made a smart choice engineering supports out of spring steel (presumably that’s what it is) which would easily give that needed ten thou of movement. If I were doing it again I’d simplify the supports a lot – maybe by using a teflon rod or a rod and simple teflon bearings. Or copy Bearpaw’s design.
 

dvd8n

Supporter
Supporter
AIUK Saviour
#5 - The Barrelled Arrow Problem

Barrelled and tapered arrows are a problem.

Basically, barrelled and tapered arrows protrude down in the middle by virtue of their shape, reducing the nominal half inch between the arrow and the central column by enough to significantly affect the spine reading.

I spent a long time wondering how Bearpaw got around this problem, and was totally baffled. Then I found an instruction manual online that contained a statement to the effect that arrows with non-parallel shafts could be measured and compared with each other but the absolute spine readings would not be accurate. So, Bearpaw hadn’t got around the issue at all, which made me feel a little better.

Now, barrelled and tapered arrows aren’t an issue for spine gauges with pointers or dial gauges as the gauges are so easy to zero that you can set the zero for each individual arrow. This doesn’t work for the Bearpaw as (I believe) it is calibrated at the factory, and a user can’t adjust it, so the problem is handwaved away. And it’s not great for my design either as my design is so difficult to calibrate. The pillar’s position is set with a grub-screw – if you loosen the screw then the pillar comes loose and drops down. Consequently it’s really hard to make fine adjustments. If every arrow that you measure is the same shape then you can spend time setting it for those arrows and forget about it, but small tweaks on an arrow by arrow basis aren’t really feasible.

If I was making it from scratch I’d thread the bottom of the pillar which would allow for easy, fine micrometer-like adjustments by turning it slightly. I think that’d be a worthwhile improvement anyway, but it’s still not an adjustment that you’d want to be doing a lot.

All in all, I’m not sure that there’s a good solution.
 
Last edited:
Top