Top Deflection rig

[jeffhigh]

Here's my take on a device for measuring top defections for monopole mobility calculation
This one came out at 22,4 x10-3 which I am happy with considering it is a seven string, with a large golpeodor plus 3 k&k Piezos

[Craig Bumgarner]

Interesting. First deflection jig I've seen other than the one I cooked up and like the simplicity of yours. I'm interested in how you are applying the load. Is there a spindle hiding behind the gauge? A picture from the side would be interesting. In my jig, things get crowded eithe the load and the gauge wanting to be in the same place.

Is the jig fairly stable sitting on the guitar without being clamped to it? I clamp mine but haven't tried it without. I do use a heavier weight, 5 kg, and the idea of the jig turning over is unpleasant.

At first I wondered if sitting the guitar on a towel would somehow "absorb" some of the load. I support the guitar with wood wedges under the lower edges but now that I think about it, I'm guessing no matter how much the guitar settles into the towel, the load is still being fully applied to the top "first". (I'm not much of an engineer). The towel is certainly easier.



I'll post a picture of my jig later today.

[jeffhigh]

The jig is sitting on top of the rims right at the edge (the support surface is slightly tapered) so it does not matter what the guitar itself is sitting on.
The support blocks move in and out and rotate and are locked by the socket head screws on the top.
It is very stable
The load sits on the small platform which has a small hole drilled in it and fits on top of the dial indicator spindle and the load is transferred to the bridge through the dial gauge spindle itself. So you are measuring and applying the load at the same point.
At the moment I just have a 295.5 Gram weight on it cause that's what I had, I'll probably get something about twice this.

[Tom West]

The jig is sitting on top of the rims right at the edge (the support surface is slightly tapered) so it does not matter what the guitar itself is sitting on.

I think this approach is so important when trying to get accurate measurements.
Tom

[Craig Bumgarner]

Here is a picture of the rig I've been using. The weight is not shown, a barbell weight goes on the spindle at the top in the middle. Except that my gauges are offset from the load point, I think it is reasonable accurate and stable. For the sake of simplicity and easier transport, I'll probably modify or make new similar to Jeff's. I like the ideas of the weight on the gauge spindle, just setting the bar on the top edges (no clamps) and the towel support. Anyone have a different version? Pictures?

I said I used a 5kg weight, not true. 5 pound, 2.2kg. My guitars are like archtops with a floating bridge and tail piece. Under string load the downward pressure of the strings is ~ 12kg, so 2.2kg is not out of line on mine and more weight reduces the possibilities of error due to play in the rig.

[Craig Bumgarner]

I've been thinking of switching to lead shot in a plastic bag for weight. Sooner or later, I'm going to drop the weight and a solid disc of metal will not doubt damage the top.

[jeffhigh]

Hi Craig,
I'd suggest rethinking the 5 pound load
Whilst this is probably close to the static string load on a selmac, the dynamic forces applied to the top from string vibration are much less, Trevor has calculated them to be equivalent to about 400 grams or about 1 pound.
There is no play in the rig with load applied through the gauge spindle.

Particularly if you are looking at testing complete,strung instruments, you would not want to double the downforce.

I'm pretty sure my rig will work on a selmac, I'll try it today and post a pic.

[jeffhigh]

Tried it on the Selmac
I needed to shim the support blocks to accommodate the height of the arch and the bridge.
I used 516.8 grams to get 0.04mm deflection
1000 grams would be better
In conjunction with an uncoupled top frequency of 250 HZ this gave me a Monopole mobility of 12.2

[Trevor Gore]

Nice rig, Jeff. I might just convert mine to something like that.

I thought I had a pic of mine in the book, but couldn't find it(!), but the second pic down here shows it. The guitar sits on wedges, one as part of the gauge arm, the other free, so fine adjustment of height (gauge centring) can be made by sliding the wedges in and out.

[Tom West]

Trevor: I'm not getting how the load is applied to the top. Looks like it's sitting on the dial indicator..????
Tom

[Tom West]

OK............The light bulb finally came on..........!! Thanks anyway.
Tom

[Trevor Gore]

Looks like it's sitting on the dial indicator...

Yes, same as Jeff is doing it. The load sits on a platform that sits on the dial gauge shaft, so that the loading and displacement measurement are exactly coincident.

[Tom West]

Some times my old brain takes a while to engage. Have been a machinist years ago and one had to be sure your dial was on a ridged support. I looked at your picture and seen the weight but said to myself" the weight is going to bend the arm the dial is mounted on". DAH One does have to keep one's mind open.
Tom

[Craig Bumgarner]

Some times my old brain takes a while to engage. Have been a machinist years ago and one had to be sure your dial was on a ridged support. I looked at your picture and seen the weight but said to myself" the weight is going to bend the arm the dial is mounted on". DAH One does have to keep one's mind open.
Tom

Well, I was thinking the same thing, that the load, if not dead center, would side load the spindle. And there isn't much for the platform to hand on to. But Tom seems to have realized something here, but I'm not sure what it is.

My idea is to have a platform with a dowel out the bottom. Dowel would be contained by a loose fitting hole in a block fastened to the truss across the guitar and sit directly above the indicator shaft. The dowel end would have a little up in it to center on the dial indicator shaft. This puts the load on the indicator shaft, but not the side loads. Course, still have to be careful to balance the load well or friction will be an issue in the measurement.

Maybe, as Tom suggests, I am missing something. Is the platform really sufficiently stable just sitting on the indicator shaft?

[Tom West]

Craig: I was just explaining that I thought the arm would bend. This was the reason for my first post. Then I realized the force was directly through the dial on it's spindle. I don't think there would be any problem with any type of side loading.
Tom

[Craig Bumgarner]

Thanks Tom. I guess what I am picturing is the platform teetering on the shaft if the load is not carefully centered. The upper shaft on my indicator is only 5.6mm long and and 5.5mm in diameter, doesn't seem like much purchase.

[jeffhigh]

My Dial indicator is about the same Craig,
I just drilled a fractionally smaller hole than the knurled cap to give a tight fit and it works well for the range of test weights I am using.
I would not want to put a 5 pound barbell weight on it though

[Craig Bumgarner]

Okay, here's my Mark II version based on a lot of the ideas presented here. Weight on the indicator shaft, support for the indicator is on the guitar edges. I can't even get close to being stable enough to get accurate reading without clamping it, but clamping is easy and eases my mind. Wider bases would probably help, but the clamps solve the problem completely.

Went to 1 Kg weight a suggested (got a deal at the local fishing tackle store on small lead sinkers, $1/pound). The sinkers in a bag will hopefully be less damaging if I ever drop the bag.

In my version, the weight platform transfer weight to the indicator shaft by a short shaft fastened to the platform. This shaft is guided by a block of Delrin. Measurements seem accurate or at least they are repeatable. Thanks for everyone's help.

One thing I would say is this is a tough measurement to take accurately. It looks like .02mm makes a difference of ~ 20% in Specific Mobility, which doesn't allow much room for error. I have a reasonably good dial indicator but notice variations of .01 - .02mm over a series of reading both in the loaded state and the return to zero after unloading. If find if I load and unload 3-4 times the reading get more consistent and then I take half a dozen and average them in my head.

[jeffhigh]

The lead sinkers looks to be a good idea Craig, I'll probably get some.
It will probably be a lot more acceptable when testing other people's instruments
Yeah reading small measurements is difficult, loading/unloading helps as you have found. I would have said variations were about 0.005mm though
My dial indicator has markings at 0.01mm and I read to the next decimal reckon I can get within 0.002

[kiwigeo]

What about a digital guage?

http://www.ebay.com/itm/1-Travel-DIGITA ... 0644704252

[Craig Bumgarner]

Stumbled across this thread today, thought I would update with my current version. Modification is in the feet which sit on the guitar top. The 10mm x 30mm x 60mm feet can pivot a little, with friction, in the slots and have small cork pads at the ends, so each foot sit on two points only. This make the whole rig very stable and I no longer need to clamps which makes it much easier to set up quickly, takes less than 30 seconds. It can still be clamped and measurements with and w/o the clamps are identical, so I'm comfortable without them.

[Trevor Gore]

Nice rig, Craig. I should update mine sometime...

[Trevor Gore]

Well, with the next Modal Tuning Course coming up (next weekend) I finally got around to updating my top deflection rig.

Here's the old one, which works fine from the measurement point of view, but is not the easiest to move into position and adjust:

Wedge top deflection rig.jpg (134.21 KiB) Viewed 50832 times

The dial gauge is on the end of that dowel because it gets used like that in a couple of other fixtures.

Here's the new one; very much a manifestation of Jeff's rig with some minor differences.

Gantry top deflection rig.jpg (98.44 KiB) Viewed 50832 times

I kept the dial gauge on its dowel, because it will still be used in other fixtures.

Dial gauge shaft.jpg (170.26 KiB) Viewed 50832 times

The main differences are the curved shoes to fit the guitar outline a little better (for more stability) and the stack of washers which are used for incremental height adjustment if necessary (inserted between the shoe and the spacer block). As it is, (first gantry jig pic, with the washers just stacked in storage mode at the top) it should fit just about any flat top steel string or classical guitar within the scope of the 10mm travel dial gauge, so the washers are there just in case....

Gantry jig end fittings.jpg (175.41 KiB) Viewed 50832 times

It seems to work well and is certainly easier to use than the wedge rig. Thanks, Jeff!

[Dave M]

First thanks to all for the development work on the jig. I have built something looking a lot like Trevor's finished product. The idea of putting the mass actually through the dial gauge is a stroke of genius.

Second I am assuming that the measurement is done on a strung up, tuned guitar? Sorry Trevor you don't seem to have defined this in the book. I assume the pictures in the thread are just illustrative, of guitars without bridges.

Third I am much concerned at the idea of just doing one measurement. Surely it is better to do several with different masses to get at a value for stiffness averaged over a number of measurements? At the least it gives you a sanity check that you haven't got a bad reading for whatever reason.
I show the measurements I did today where I roughly cover the expected force by a bit on either side. I also repeated the experiment a few times to check that the loading and measurement mechanisms were working properly

I now have to try and get my powers of ten right for the subsequent calculations!

Dave

[Trevor Gore]

Second I am assuming that the measurement is done on a strung up, tuned guitar?

Correct. In my pics I used that guitar just because it was to hand. The majority of my guitars have fretboard overhangs, so sealing the sound hole is a pain, so this one didn't have a neck on either. The monopole mobility prior to the bridge being fitted is very different to the MM afterwards, but it's the latter one that matters.

Third I am much concerned at the idea of just doing one measurement. Surely it is better to do several with different masses to get at a value for stiffness averaged over a number of measurements?

I usually use two different masses. But multiple measurements with the same mass averaged or different masses and taking the gradient of the regression line is fine also.