DIY Thiele/Small measurements ??

[youyoung21147]

Hello,

I have ordered some Supravox 165GMF drivers, am waiting for them, and already have the data about them (Supravox gives a personal T/S parameters sheet with each driver).

The problem is that the params are quite far from the published ones (http://www.supravox.fr/haut_parleurs/165_GMF.htm) :

Driver 1
Fr : 78,48
Vas : 27,4
Qms : 7,167
Qes : 0,488
Qts : 0,457
Re : 5,5
Sd : 165
Bl : 5,6543
Mms : 5,7500


I modelled the enclosure again and it definetely affects the tuning.

I suppose (I hope) that this difference is mainly explained by a lack of break-in :

So, I plan to measure the T/S after the drivers are broken in.

I did a lot of search on the forum & the web and found that I need several things :

- a microphone, which I have
- a soundcard, which I have
- a mass to add on the driver, which I don't have
- a sealed box which I can build
- a software which I don't have


So I ask you, experienced DIYers , what you suggest me concerning :

- the mass : I can get a 0.01g precision at school, but don't know which mass I should add. Any suggestion ?

- the software : I found sample champion and Praxis, both are not free and Praxis needs quite boring cables. Do you know a good freeware program that could do the same ?

- the box : should it rather be a sealed enclosure of a given size, or a kind of open baffle ?


Any suggestion welcome !

 

[Josephjcole]

As far as mass blue-tack (fun-tack, sticky-tack) works, and is easy to add and remove from the cone. I'm not sure about your question for the box, what are you measuring using the box? Not Vas I assume. If this is for accustical measurements, I would just get all your T/S parameters, model and build your box, then do accustical measurements in that. Or maybe the box is for getting the T/S parameters, in which case open baffle is fine. Really anything to hold the driver off of the ground is fine.
Joe

 

[Geek]

WinISD can do this too.

There was another, but for the life of me, I can't remember the name 🙁
(somebody help me here - light grey GUI, almost looked "GTK-like")

 

[youyoung21147]

Thanks a lot guys !!

I'll get some very very precisely measured blutak to put on the speaker (hope I won't damage the dust cap ).

I will try WinISD for this when I have more time !

 

[youyoung21147]

I downloaded WinISD, and I found nothing to measure T/S params, in the old and new version. I looked at the help file but it seems that there's nothing like that. Am I wrong ??

 

[Geek]

In the alpha version, it does a lot of calculating of the missing stuff. If you can find your "Q's" and Vas or Fs, it'll pretty much do the rest.

 

[planet10]

To calculate Vas (& some other parameters) you require either added mass or a box of known volume small enuff to dramatically raise Fs.

I use FuzzMeasure (& an auxiliary spreadsheet) to calculate T/S parameters. Works a charm.

dave

 

[Josephjcole]

I use FuzzMeasure (& an auxiliary spreadsheet) to calculate T/S parameters. Works a charm.

dave

spreadsheet is a good idea. I've just been using a calulator, kind of a pain.


Joe

 

[ergo]

You can try Praxis from http://libinst.com/praxis_downloads.htm

Without hardware it works in freeware mode but has a full T/S parameter measurement / calculation regime available.

Ergo

PS. Oh yes - it makes a T/S parameter caclulation with curve fitting procedure, so it's much more precise than hand calculations.

 

[planet10]

Without hardware it works in freeware mode but has a full T/S parameter measurement / calculation regime available./QUOTE]

Have you actually gotten it to work. I've been talking to some people with the full-blown version and they are having (were having?) a hard time getting it to work.

FuzzMeasure is using a method based on the manual method, but curve-fitting is in the works.

dave

 

[DavidH]

Has anyone used the "Dayton Woofer Tester 2"? (It is not just for woofers, it can test any driver.)

http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=390-802

At $160 it is not exactly cheap, but I would be willing to pay that price if it worked well.

/Dave

 

[youyoung21147]

I looked at Praxis again, it seems to have nice T/S measurements. It isn't an easy soft to use but has many possibilities.

I'll try to play arround with it when I have more time, and when I have a 0.01gr precision instrument in my hands 😀


I went on Fuzzmeasure website but the download doesn't work.
Talking about this, what additional worksheet do you use with it ?

 

[richie00boy]

Looking at the list of equipment that you do have, you might be interested in guide I have written. You don't need any added mass 🙂 email me if you want me to send you the PDF and are prepared to offer some feedback on the processes. It is written with beginners in mind but assumes a (very) small amount of electronics knowledge.

 

[planet10]

I went on Fuzzmeasure website but the download doesn't work.
Talking about this, what additional worksheet do you use with it ?

Chris will be interested that you failed to download it... can you try again and let me know, Note: that to use FuzzMeasure you have to be running OS X and given that you are trying out Praxis, my guess is that you aren;t.

I've attached the spreadsheet, which i downloaded & modified for my use from "Best ever TS parameter calc.xls" (i measure reams of drivers). Hopefully it is self-explanitory (only calculates using the added mass method)

dave

 

[Josephjcole]

Dave,
thanks so much for sharing your worksheet. After reading your ealier post I thought about making my own spread sheet, unfortunatly inertia was against me🙄 . I had always done the calculations with a calculator, which was a real pain with all the conversions and what not. I always had to calculate it at least twice to make sure I had not messed anything up.
Thanks
Joe

 

[youyoung21147]

Yes, I'm not running Mac OS but winXP.

It explains the download problem.

Thanks for the worksheet anyway !

 

[planet10]

It explains the download problem.

What OS you have should have no effect on downloading (you'd just have problems running it 🙂)

dave

 

[youyoung21147]

yes, I explained it bad : I could download the file but couldn't open it 😀

 

[Geek]

Works great on Linux and OpenOffice 🙂

Thanks Dave! 😀

 

[CrisTUFR]

FuzzMeasure is using a method based on the manual method, but curve-fitting is in the works.

If you guys are interested, you can find more details of my curve fitting work here:

http://www.supermegaultragroovy.com/FuzzMeasure/perfect_fit.html

It's a pain to implement, but now that I have it working, it's only a matter of time before it hits the final product.

Cheers,

Chris