Site Links
Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
Quick Links
Categories
In this Discussion
Who's Online (5)
- charlielaub 3:44PM
- Eggguy 3:44PM
- meniscus 3:44PM
- rjj45 3:44PM
- Tom_S 3:44PM
Driver testing stand
I just finished building a heavy duty driver stand/vise for T/S parameter testing. Very stable (not tippy). It can clamp big as well as moderately small drivers firmly in place for all tests.
Cost:
1) Two eight foot long 2x4's: ($15)
2) Tool Shop Ratcheting bar clamp ($10), 3) Heavy duty 3.5" door hinge ($2.50)
4) Carrage bolts & lag screws ($5) .
Total: ($32.50)
Design goals:
1) Quick & safe clamping & removal.
2) Clamp to magnet, not flange.
3) Horizontal mounting orientation.
4) No blocking of vents or cones.
5) Stable with heavy woofers (25 lbs).
6) Wide clamp adj. range (2" - 7.5" dia.)
7) Not designed for drivers < 2" dia.
Here is the finished stand/vise.
More pics & details to follow
Comments
You must be retired!
If these prices keep going up, I'll have to come out of retirement and get another job!! I can't believe two 2x4's now cost over $15. Last time I bought 2x4's, they were a couple bucks each!!!
Here is the stand with a big, heavy 12" Titanic MK III mounted. Very stable and secure. Ready for testing. 18 lbs with a 6.5" diameter magnet.
Retired and with space to spare...FWIW I just hold drivers in my hand for impedance and T/S. Big heavy ones I'll prop up on some blocks.
This time with the GRS 12SW-4HE mounted. I'm getting ready to build a pair of JR's Twin subs.
I've noticed that I get slightly different measurements depending on how I hold the driver. That is one of the reasons I built this. Another reason is that, when hand holding, I keep knocking the alligator clips off. And it looks cool. And much cheaper than a Klippel Pro Driver Stand!!
Here is the stand with a Peerless DA25BG08 mounted. Notice how the bar clamp remains in perfect alignment, even with small drivers. Driver magnets are pressure clamped at 3 points, regardless of magnet size.
Here is the Tang Band W6-1139SIF mounted (medium sized driver):
Here are a few more close-up details of the design:
The vise is hinged on the left side using a standard 3.5" door hinge:
A small rubber strap tie (from MWAF grab bag) holds the top clamp jaw in place.
I used a dremel to grind angled pivot surfaces for the top clamp jaw. As you tighten or loosen the vise, the clamp bar slides along a small notch cut into the end of the 2x4, keeping everything in perfect alignment.
The 16" x 18" base is made from 2x4's screwed together with lag bolts and cross braces:
The lower bar clamp jaw fits into a custom machined notch that I cut into the 2x4 using my jig saw and dremel:
When the two lag bolts are tightened, the lower clamp jaw is tightly clamped in place and cannot move.
Nifty!!
... didn't realize wood went back up.
2x4x8's were $15 here for a while, then down to $3, now back up to $7
Bought eight 2x4 last Saturday to build a temporary solar array mount. $120 for the lumber. You got two for $15, that's a lot less than I paid.
I just hope the price of new cars comes back down. What we really need now is a huge glut in the automarket so that I can re-stock my driveway with new vehicles.
I guess I haven't shopped for lumber in a few years. Those prices are crazy! But with the post-Derecho repairs still going strong here in town, I've been avoiding projects that call for lumber until everyone is finished with their repairs.
It usually does after a rest and further encouragement . . .
I conducted a small comparison test of three different free-air T/S measurement techniques using one of my new GRS12SW-4HE subwoofers. The three techniques were as follows:
1) Woofer resting face up on two short 2x4 stubs. The pole piece vent holes were 3.5" above the carpeted floor:
2) For the 2nd technique, I hand held the woofer by the magnet, careful not to block either the main pole piece vent hole or any of the other smaller vent holes on the back of the woofer:
3) For the 3rd technique, I mounted the woofer by the magment in my vise, as shown in my previous photos above.
For each technique, I made 5 measurements using DATS V2. In this way, we can isolate the normal measurement variations from the variations caused by the technique itself. I did not bother making VAS measurements, as I wanted to keep things simple. Here is a spreadsheet showing the results:
The main difference that strikes me is the Qms measurement. The set of 5 gray floor measurements show a Qms of about 4.08. The set of 5 yellow hand held measurements show a Qms of about 4.45. And the set of 5 blue vise measurements show a Qms of 4.67. To me, this indicates a mechanical difference, maybe due to the difference in gravity pulling down on the driver. The hand held measurements were made with the woofer pointing up slightly, about 60 degrees or so, because it was difficult to hold it at 90 degrees like the vise.
Another interesting observation is the variation in Re for the hand held measurement set. You can see that it varies quite a bit compared to the floor or vise measurements sets. This is probably due to movement of the alligator clips because my hand was moving around during the measurements.
Also, I did a couple of memory overlay graphs in DATS V2.
The first one compares green Mem 1 (floor), with purple Mem 12 (Vise). The floor measurement has a glitch below 20Hz. I am guessing that this is probably due to the vent holes being 3.5" from the floor. What else could it be?
The 2nd overlay graph compares green Mem 6 (hand), with purple Mem 12 (vise). These two curves appear almost identical. The difference in Fs is only 2% and you can barely see that in the graph.
Conclusion: Not much difference between floor, hand held, or vise. But I am still glad that I built the vise. It makes it much easier to test large drivers.
I wonder if that little step below 20Hz is the carpet flexing just bit at that frequency. The carpet's Fs!
How much do the measurement's from day to day? Dats is not what I would call a repeatable measurement system.
I have not noticed short term daily variations using DATS V2, but I have not been testing for this. I have noticed significant, longer term, 3 to 6 month variations on other drivers. I've always written this off as due to seasonal changes in air pressure or relative humidity. I'll wait a couple days, re-test this driver in exactly the same way, and then post a new spreadsheet for comparisons. The driver has not been broken in or playing music at this point, so this should provide a good "day to day" type comparison.
Interesting. The carpet has 1/2" thick foam padding, so there might be something to this. Also, when I jump up and down at the spot when this test occured, the floor joists vibrate just a tad as well.
I've seen bigger changes than what's shown above from a couple deg difference in ambient temperature.
The real question is - with all that work that you put into building a measurement jig, what are you still doing using a DATS ?
Force of habit more than anything else. And the Arta jig is right next to my laptop with all the cables still connected. All I would have had to do was plug the
TRS into the UR22 instead of plugging the DATS into the laptop. Go figure. So, when I run the measurements again, I'll also run up and post a set using the new Arta jig.
John, today I ran the same measurements again on the same driver using DATS V2. Temp in my house is the same as before, about 70F. HVAC cycled off for all tests.
Humidity is probably just a tad higher today, as it has been raining on and off all day. Measurements are almost identical, so my DATS appears to be fairly stable on a daily basis. Not sure if this answers your question. See new spreadsheet below. Are you using the original DATS or the V2 version?
Dcibel, I fired up the Arta jig and re-ran the same tests comparing floor, hand held, and vise measurement techniques. As you can see in the spreadsheet below, the results are very similar to my DATS V2 tests. I didn't bother measuring VAS, as I wanted to keep things quick and simple. Note that the driver Re of 3.64 ohms was measured with my Ex Tech 530 meter, which can resolve to 0.01 ohms with an accuracy of +- 0.3% of reading + 9 digits. This Re measurement was made with the driver clamped in my vise (not on the floor), because the Re value bounced around too much when I placed the driver face up on the floor. The big woofer cone acts as a microphone and really messes up the measurement when placed face up.
I was curious as to what was causing the small glitch at 19-20Hz during the floor measurements, as this glitch completely disappeared during the hand held or vise measuremnts. Tom speculated carpet Fs and I think, based on additional testing, that this is correct. The glitch shows up during the floor measurements for both DATS V2 and Arta jig tests. Here is the LIMP screen shot showing the same floor glitch:
To test Tom's theory, I clamped the woofer in my vise and tipped the entire vise onto it's back. I placed two 4 inch diameter PVC couplers under the 2x4's to hold the magnet about 3.5 inches above the carpet. This eliminated the need for the two small 2x4's stubs used in previous tests.
I ran the sweep again, and this time the 19-20Hz glitch completely disappeared (see graph below). The glitch was being caused by the relatively lightweight 10 lb speaker deflecting up and down on the carpet. When clamped into the vise, the extra 30 lb mass of the vise plus the 10 lb mass of the speaker damped out the carpet Fs.
Bill can you post the data instead of a picture? I'll calculate the accuracy.
100% of the repeatability problems I had when I owned a WT3 was with the shitty alligator clips. It’s otherwise just a cheap USB audio codec with a sense resistor built in.