T-amp for testing and design?

[littlesaint]

Anyone use a T-amp for doing measurements of drivers?

[Nemophyle]

i don't know but you could be the first and tell the world if it is viable or not.

Basically if you do a loopback (plug the out of soundcard on the in of the t-amp , and outs of your amp on the in of your soundcard. REmember to start at 0 output so you don't blow your soundcard, even if i don't think a T-amp has enough guts for that) and measure the Frequency response and harmonic distortion.

Then post your results here if you can't tell by yourself if it's viable for driver testing

BTW, the t-amp should be ok i think. Maybe you can find theses curves somewhere. WIth all the t-amps fans i'd be surprised if you couldnt find that on diyaudio or somewhere else

[littlesaint]

From the reviews I've read, it looks like FR is flat from 100Hz to 10KHz, and with some easy mods you can flatten it out some more. THD gets pretty high if you crank things up, but for simple testing that's not necessary. Looks like you'd need a good wall wart with clean power.

[cjd]

Zaph has nixed the T-Amp as viable for use in a measurement rig.

C

[Quwiksilver]

There's a lot more to t-amps than the puny little Sonic Impact that I believe you're referring to. Try:

41hz.com
general comparison

Some of those on 41hz.com not based on the Tripath 2020 chips can put out quite a bit of power.

Scott

[JonP]

Been there, done that, luckily didn't blow the amp or my soundcard input...

NO, you can't use a Sonic T amp for this, due to it's "bridged" type of output stage. The two speaker connections are at 1/2 of the supply voltage, and you will create a short when you connect one side to ground in the measurement hookup. (the input ground will be the return path)

I thought it would make a wonderful small portable package... just can't do it with this amp...

Apex jr had some little 10W (non-bridged) amps surplus from some gear... think they were $10 or so... I've used one of them.

[Mazeroth]

I tried and tried to get accurate impedance measurements with SoundEasy and was ready to give up. I then switched to an old HK amp and voila, perfect impedance measurements. Out of curiosity I then tried a Panasonic XR-25 digital receiver and got the same results I did with the T-Amp. So no, for speaker building you should not use a T-Amp.

EDIT: At first I was using the T-Amp. Sorry for not being clear.

[fjhuerta]

I used the T-amp with JustMLS. I do my impedance plots with the Woofer Tester 2. The T-Amp worked wonderfully. It worked even better when I modded it with bigger stiffening caps and fixed the input RC circuit with polypropylene caps - bass was *significantly* better - the T-amp has a roll-off in the bass. If you want to fix it, you need to try this mod.

[ch83575]

I am not great with electronics, so I might be totally off base here, but I thought that the sonic impact t-amp used only a positive leg for power, so it has some kind of elevated ground voltage in the speaker loop. Ok for some things, but not if you are using a reference probe. Then you need some sort of differential probe or something like that. Again, I am just trying to regurgitate something I remember hearing... I dont have an full grasp of all of the implications, so take it with a grain of salt .

-Chad

Sorry... must have missed jonP's post... yeah, what he said.

[littlesaint]

I think I'm just going to go with a BPA-1 from Emotiva for a test amp. They're on sale this month, and I'll get more all-around use from one of those anyway.

The T-amp looks interesting as a concept, but I'm looking to build speakers right now, not amps.

[JonP]

I used the T-amp with JustMLS. I do my impedance plots with the Woofer Tester 2. The T-Amp worked wonderfully. It worked even better when I modded it with bigger stiffening caps and fixed the input RC circuit with polypropylene caps - bass was *significantly* better - the T-amp has a roll-off in the bass. If you want to fix it, you need to try this mod.

Yes, they do roll off a bit by 20Hz, with the stock input...

I'm interested in your setup, and how you avoided the floating speaker lead/short problem. How does JustMLS use the reference or sample leads?

The problem with ANY bridged type output, not just the T-amp, is both speaker leads are floating above ground. Usual setup is, you have your sound card input sense leads attached to some "hot" side of the amp output, and the shield to the "cold" sode. Normally, this is system ground. In a T-amp or other bridged amp, it is floating at 1/2 the DC supply.

You also have the same shield ground of the sense lead connected thru the sound card and cable to the card output... going into the T-amp input. T-amp input is system ground... and if you followed all that you see there is a path from amp out - to the ground, and you have a short.

I don't see a way to avoid that, except by not connecting the shield ground of your probes/sense leads to anything, which might allow noise to get in the measurement, or by putting large caps to AC couple the output or input, which might influence readings as well... Maybe I've missed another way...

After all that, I re-read your post and noticed you're using it with Woofer Tester. ops: Maybe there isn't external sense connections and the amp can 'float" in this case?

[fjhuerta]

Yes, they do roll off a bit by 20Hz, with the stock input...

I'm interested in your setup, and how you avoided the floating speaker lead/short problem. How does JustMLS use the reference or sample leads?
...
After all that, I re-read your post and noticed you're using it with Woofer Tester. ops: Maybe there isn't external sense connections and the amp can 'float" in this case?

You are right!

I did use a probe in order to do two channel measurements, and floated the ground probe. It worked fine.