Amp module smoke and noise.

[dkelly60]

I have two problems that I hope I can get some help with.
I am in the process of updating a 3 way active system I built about 14 yrs ago.
The original setup used Audax 8", 4", & 1" drivers with LM3876 chip amps and active crossovers. I developed a problem with one 8" driver and could not get the same audax anymore. So I decided to update the whole system. I was hoping to use Class D amps and a miniDSP as the crossover.
I am currently using these TDA7498 modules 20VTO DC36V TDA7498 100W 100W Class D High Power Amplifier Board DC | eBay
[IMGHTTPDEAD]http://pcsmarts.biz/sites/default/files/amp.png[/IMGHTTPDEAD]
It's been a long time since I did any electronics, so please excuse my ignorance.
I wanted to check the noise and gain by looking at the output of the amp so I hooked up a 10 ohm 5w resistor to measure with an oscilloscope. To my surprise the amp smoked and killed the output on that channel within 30sec, why would this be?
Second problem is I am disappointed at the noise from the power amp (hence wanting to look at output). It is particularly noticeable on the tweeter and can be heard at 3-4 meters. Are there any mods that might help reduce this noise?

 

[wwenze]

Did it die *only after* the oscilloscope connection?

Shorting oscilloscope ground to the negative output terminal of a bridged amplifier is not a good choice.

See also:
http://www.youtube.com/watch?v=c7dV1PNPxAk&t=1m34s

Noise... TBH I feel that film caps aren't needed to filter out HF in the power supply - SMPS rarely has film caps on the output and they switch at a few octaves higher than audible frequencies. A good low-ESR electrolytic is sufficient. And the capacitor on that amp (that lone blue one) does not look like a good low-ESR electrolytic. It also looks hugely undersized - even the Lepai LP-2020 one looks bigger... and that's before factoring in the Lepai one is only 15V while the one in the TDA7498 amp would need to be 35V

http://i1137.photobucket.com/albums/n513/menacewebpics/lepai/1.jpg

Any mods? Well... how about changing... all the parts? The film caps at the output look weak as well so there may be insufficient filtering there. But first I would find a good quality DC power supply (Anything from 15V to 32) and then upgrade the capacitor on the amp board to something legit like Panasonic FM or Rubycon ZL. Some people may want more capacitance, but try to fit 2200uF first... or even 1000uF.

This post shows how big Panasonic FM 2200uF 25V are, when being used to mod the Lepai
http://www.diyaudio.com/forums/class-d/90500-lepai-t-amp-ta2020-93.html#post2764716

You can try the Panasonic M series, they are smaller for the same capacitance. Trade off being higher ESR but still useable for SMPS.

 

[alarrr]

Hi got same one from ebay, I was testing how much can chip take before thermal or cliping protection starts. Results are in tda7498 thread with a picture

 

[rayma]

I hooked up a 10 ohm 5w resistor to measure with an oscilloscope.
To my surprise the amp smoked and killed the output on that channel within 30sec, why would this be?\

The amplifier outputs are bridged and floating. The scope probe common clip is grounded.
You would have to use a differential voltage probe on the output terminals to measure this amlifier without shorting one side.

 

[Pafi]

Those TDA ClassDs have quite high noise. If your tweeters have high sensitivity, then you must choose an other amp, like TPA3116 or TAS...

You may try to set lower gain if it is too high by default.

 

[alarrr]

Also why are there 2ohm resistors between speakers? Woudnt it those reduce amp power?

 

[alarrr]

Also why are there 2ohm resistors between speakers? Woudnt it those reduce amp power?

Those white boxes you see on the picture, as i understand it whole point of using large diameter wires in audio wiring is to keep resistance down. Now add 2 ohm there, what=?

 

[theAnonymous1]

Also why are there 2ohm resistors between speakers? Woudnt it those reduce amp power?

Class-D output Zobel.

 

[dkelly60]

Did it die *only after* the oscilloscope connection?

Shorting oscilloscope ground to the negative output terminal of a bridged amplifier is not a good choice.

Did it die *only after* the oscilloscope connection?

To be honest I don't remember if it smoked before I hooked up the oscilloscope, I think it was most likely after. Thanks for the links and tips.

You would have to use a differential voltage probe on the output terminals to measure this amlifier without shorting one side.

Thanks I will look into this.
Obviously I need to do some reading on class D

 

[dkelly60]

I have been looking at this a bit more I sorted the smoke problem and have checked the voltage gain its 50 which also matches what I see on the multi-meter. I can see a 25mv pp signal on the output at 330KHz, would this be the audible noise source?
I have also looked at this article Trevor Marshall - Class D Audio Amplifier Design - TDA7498 Output filters
It is the same chip in a Sure module, since I don't have a circuit of the module I am using could I have some feedback on the relevance to my situation?

 

[alarrr]

Class-D output Zobel.

No this 2ohm is not zobel. Its not between +and -(and capacitor), its only on - line.
Yea mine has also hiss. But its only audiable max half meters range.

 

[alarrr]

 

[theAnonymous1]

Ah, I see. So that resistor is in series with the output? Strange.

The entire output section of this amp looks wrong according to the datasheet. Those inductor values are too high (22uH recommended) and there doesn't appear to be any zobel components at all.

I see other cheap TDA7498 boards that have done things correctly, so I don't know what the designer of this board was thinking.

 

[dkelly60]

The white 2R2 resistor has blown on the dead channel. The unit might work if I replace it. It is a 2R2 1% 5 watt but I don't know where to find a replacement. Everything I come across is pigtail and 5% tolerance, any ideas?

 

[rayma]

Thanks I will look into this.
Obviously I need to do some reading on class D

Some more conventional amplifiers are also bridged, with the same problem.
Even Audio Research's 40 year old tube amps have the 4 Ohm taps grounded,
instead of the 0 ohm taps, to allow partial cathode coupling.

 

[wwenze]

The 2.2R is series with -ve? Probably a very bad attempt at allowing the -ve terminals to be connected together.

Most chipamps have protection against shorted outputs so replacing that might work. Or you can bypass it altogether.

 

[tiefbassuebertr]

I have two problems that I hope I can get some help with.
I am in the process of updating a 3 way active system I built about 14 yrs ago.
The original setup used Audax 8", 4", & 1" drivers with LM3876 chip amps and active crossovers. I developed a problem with one 8" driver and could not get the same audax anymore. So I decided to update the whole system. I was hoping to use Class D amps and a miniDSP as the crossover.
I am currently using these TDA7498 modules 20VTO DC36V TDA7498 100W 100W Class D High Power Amplifier Board DC | eBay
[IMGDEAD]http://pcsmarts.biz/sites/default/files/amp.png[/IMGDEAD]
It's been a long time since I did any electronics, so please excuse my ignorance.
I wanted to check the noise and gain by looking at the output of the amp so I hooked up a 10 ohm 5w resistor to measure with an oscilloscope.
To my surprise the amp smoked and killed the output on that channel within 30sec, why would this be?
Second problem is I am disappointed at the noise from the power amp (hence wanting to look at output). It is particularly noticeable on the tweeter and can be heard at 3-4 meters. Are there any mods that might help reduce this noise?

At that time occurs the smoking? Already with the resistor itself (then it could be cause the parasitic inductor in case of a winding version)? Or after connecting the oscilloskop (bridge amp configuration is GND-free)?
If no of this is the reason for the unwanted damage, check the schematic diagram of your module and compare it with those from the datasheet on page 10 - go to
http://www.st.com/web/en/resource/technical/document/datasheet/CD00244535.pdf
- maybe this will provide the right advices.

Concerning the background noise check my threads in case of the TDA7492/SMSL-SA-50 - basically the same issue.
http://www.diyaudio.com/forums/class-d/286090-noise-smsl-sa-50-tda7492-low-level.html
http://www.diyaudio.com/forums/class-d/286090-noise-smsl-sa-50-tda7492-low-level.html

 

[alarrr]

I wonder what would happen If removed those white resitors and connected directly