diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Class D (http://www.diyaudio.com/forums/class-d/)
-   -   Strange IRAUDAMP1 problem (http://www.diyaudio.com/forums/class-d/122047-strange-iraudamp1-problem.html)

theAnonymous1 26th April 2008 03:27 PM

Strange IRAUDAMP1 problem
 
I have two IRAUDAMP1 boards that where made by an unknown individual. The layout looks nearly identical to the evaluation board I have. One of the boards is acting very strange and I lack a scope to try and troubleshoot.

On the bad board, if the frequency of both channels is not the same the output devices start to make noises (oscillate?) and get VERY hot. If I don't quickly re-match the frequencies the outputs self destruct.:hot:

If the frequency of both channels are matched the amp will work but the MOSFETS still get really hot. Also, when switching the amp off/on sometimes there will be no output at first. I found that if I "tap" on the woofer cone the amp will come alive, kind of like a kick start. Very strange.:confused:

The other board I have is identical an works fine. Like I said, I lack the equipment to do any advanced troubleshooting, so if someone has any idea of what the problem could be I would appreciate the input.

lumanauw 26th April 2008 05:28 PM

It's very difficult to troubleshoot classD amp without an oscilloscope.

Tekko 26th April 2008 08:01 PM

You mean impossible? ;)

theAnonymous1 27th April 2008 12:19 AM

Quote:

Originally posted by lumanauw
It's very difficult to troubleshoot classD amp without an oscilloscope.

Quote:

Originally posted by Tekko
You mean impossible? ;)
I figured as much. I guess I was just hoping someone would have an idea what it could be.

I'll just shelve it for now until I can afford a scope.:o

Corax 30th April 2008 02:05 PM

@theAnonymous1:

1. If the MOSFETs are getting (really) hot it's kinda problem of a too low dead time - for a (very) short amount of time a shoot through current will flow because i.e. Q5 & Q6 will conduct at the same time.
Therefore the following question arises: Are the original output MOSFETs used (IRFB23N15D) as listed in the BOM?

2. If both amp channels are not in sync it's more likely to see (hear) an effect called power supply pumping. This just happens with single-ended PWM output stages driving reactive loads. The reactive current will flow back to the power supply and pushes the voltage to higher values. This however will usually extensively happen when driving loads at high power rates and with low output frequencies (i.e. when used as subwoofer amp). The only (cost effective) way to lower that phenomena is to use laaaarge bulk caps. Since you haven't mentioned the power supply I expect small bulk caps of maybe 4,700F. In this case you better use at least 10,000F or even 22,000F or more.
I experienced that phenomenon myself and got rid of it by either using completely separated power supplies - however you might use one transformer but with separate windings for each amp channel - or you synchronize both amps and invert the incoming signal to one amp and invert the output connectors of this channel. This way the reactive currents will not arrive at the same time with equal direction at the power supply bulk caps and won't add up.

3. If you must 'tap' on the cone of the speaker to force the amp to operate sounds like the amp is not self-oscillating until you inject a voltage to the amp - the speaker also acts as a microphone.
In this case it's really hard to say where the issue can be found.
A good starting point however will be Q9 (Q10 respectively for CH2) which form the so called 'Startup Circuit' as described on page 7. Check also for the recommended 'Power On/Off Sequence Timing' on page 9. If that's Ok so far check the mentioned bootstrap circuit/capacitor (C23 and C24 respectively). Check also if the MURS120 (D14 and D16) for shorts or open circuits.
Not to mention to check all solder joints if they look good - otherwise resolder them with fresh solder.

Basically a DSO would be quite handy when checking waveforms and timing criteria.

Don't forget to post your progress. I'm very keen to figure out what was/is wrong with the older design of the IRAUDAMP1 because I'm at the point to decide if I build the IRAUDAMP1 or the IRAUDAMP3.


All times are GMT. The time now is 12:00 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2