Ransom
Ransom UltimaDork
9/6/19 7:13 p.m.

I'm surprised I'm not finding more info on the web about this, or how dodgy what little info I'm finding is. Seriously, I think Hot Rod had an article about a "shock dyno" that involved dropping a weight on it and seeing how far it moved. sad

I had a cursory google for this having come up on GRM before and came up empty.

Like the title suggests, I'm curious about DIY shock dynos, and wonder whether there's any info out there about how to go about it.

I figure the worst case involves doing some research to figure out what sort of forces and velocities we're dealing with, working out power demands from that making an educated guess about how much travel is needed, and sourcing sensors for displacement and force, and finding some sort of system for logging.

It's getting ahead of myself, but I kinda figure it'd be good to figure out how to do rapid force reversals and not just a sine wave or whatever you'd get with a simple crank (or I guess a Scotch Yoke is often used, which I think mostly means the shock's motion really would be sinusoidal rather than having the different timing through bottom and top of travel); the latter would cover force/velocity, but knowing how much "float" a damper has, or generally what behavior it exhibits at the sort of frequencies you see over bumpy ground would be good to know... Maybe with the right motor and driver you could just do that programmatically, though that seems like it could quickly get more expensive than the bare power demands would require.

Aaaaand I'm already rabbit-holing.

Anybody have any links or thoughts or... ?

_
_ HalfDork
9/6/19 8:14 p.m.

This diy would be nice. Especially with How often people recommend shocks based on “ how it feels”, which is usually just a euphoria to justify the absurd money spent on a name. 

djsilver
djsilver Reader
9/6/19 8:18 p.m.

I've read a few discussions about DIY shock dyno's and decided that if I want to dyno shocks it would be cheaper and more useful to send them to someone with a real shock dyno.  

Ransom
Ransom UltimaDork
9/6/19 8:24 p.m.

In reply to djsilver :

That depends on whether you want to muck with them and see what changed, or map a range of adjustments to see how much each quantum of adjustment affects the result. Or have funny form factors (e.g. old British cars with lever arm dampers).

I suspect your point is "on point" for the case that you want to know how Damper A and Damper B (or Damper K and Damper B) compare.

freetors
freetors Reader
9/6/19 9:09 p.m.

I've seen some fsae guys out together some really nice and relatively inexpensive damper  dynos. Apparently the electronicals are pretty cheap and straightforward if you happen to be a whiz with that kind of stuff. You could probably get by with a load cell and a linear pot with very high baud rate. They could possibly feed into an arduino or Matlab or some such thing.

The next big showstopper is you'd practically have to have a machine shop at your disposal to make all the adapters you'd need. The tolerances would have to be very tight since any slop would show up on your dyno graphs.

This stuff makes sense for someone whose job is shocks, but for anyone else it's a lot of work that's hard to justify.

weedburner
weedburner Reader
9/6/19 10:07 p.m.

Instead of outfiting a dyno, outfit your car with the sensors/logger and turn it into the dyno.

 

Grant

 

Nate90LX
Nate90LX Reader
9/7/19 8:22 a.m.

If you want accurate data your shocks, then you should investigate the cost of paying a shop to dyno them vs the cost to build your own dyno.  Accuracy and repeatable day is very import in shock dyno charts.

Also, what is the point of dynoing the shocks?  The dyno chart will give you data,  but it’s difficult to understand how that transfers to the reactions on the vehicle. 

You will be better off spending time on understanding how to tune shocks and accurately measure all the parts (disks, pistons, rods) to make sure they are consistent. Then go drive the vehicle and see if you can feel the change. Then do that at least 10 times or more. That is how you tune shocks. If you were doing it professionally, you would dyno each change, but that is really only used to confirm you built them properly and each pair reacts similarly. 

You build something to give you data, and they might be fun. But without accurate and repeatable data, it’s not worth much. 

Ransom
Ransom UltimaDork
9/7/19 1:29 p.m.

Let's say I have archaic lever arm shocks and I want to compare them to modern, good-quality shocks, that I want to tweak the internals of my lever shocks, and I want to see the changes in a quantifiable manner.

Let's also say that I'm more likely to get machine shop time than track time on a regular basis.

I think the concerns about building something solid and repeatable are valid but not insurmountable.

Datalogging on the car is great, but without a load cell on the damper mount, separate from the spring mount (so, awkward on lever arms, *really* awkward on coilovers), you're not going to isolate the damper's forces. Unless I'm missing something? In any case, I can't really A/B shocks when B won't fit on my car without reconstructive surgery, and certainly not in a way where you'd want to jump back and forth.

Now, datalogging position and velocity on the car so you can correlate to behavior, *that's* something that would go nicely with data from mapping the dampers.

I'll see if I can find anything with the FSAE clue.

Ransom
Ransom UltimaDork
9/7/19 1:32 p.m.

Interesting: http://sportdevices.com/shock-absorber.php

EDIT: This place has instrumentation, kits, plans...

Our Preferred Partners
wQejy2JLg4texySOJ1ZMwwr2K4oF5UFIOqEMSF4XdoVHeX6HJl2Lt6pOzeVaACgn