Okay so i got this amp from a friend of mine, he bought it off the internet as a repair. The amp came with blown supply fets and a few blown output fets. He does not know much about electronics but i can see he replaced all the output fets, powersupply driver trannys and the output stage gate resistors as well. The amp got to me with a problem described as, it is louder at some points and it is behaving strange. He said some of the output fets(irf640) warm up more than the orthers. So after a few measurements i found out that two of the gate resistors on one side were swapped. After swapping them around one side seems fine, and they did heat up at the same time. So going to the other side of the amp something must be horribly wrong, this side is heating up like a champ, and comparing it to the good side did not help at all. At this point i have removed the pre driver bjts and if i fire the amp up the "good" side of the amp does not heat up at all and if i check their drain it is around (with the other sides drivers connected i was seeing 63khz) 310khz which seemed ok to me.
Could someone help me a bit? A schematic would be nice, but any help is greatly appreciated. I am just coming from the analogue amplifiers world so repairing a D class is a bit new to me. I do not understand how could this happen. If anybody has some great info to read about d class amps i would greatly appreciate that too. Thanks again
Csutak Gergo
Could someone help me a bit? A schematic would be nice, but any help is greatly appreciated. I am just coming from the analogue amplifiers world so repairing a D class is a bit new to me. I do not understand how could this happen. If anybody has some great info to read about d class amps i would greatly appreciate that too. Thanks again
Csutak Gergo
In the future, please use the 'Go Advanced' and 'Manage Attachments' options to post images. The threads are here to help others as well as the OP. Posting the images here ensures that they will be available for those who may need the information in the future.
The two 1D transistors in the center of the board run hot and fail often. Those get replaced, even if they check as OK with a meter.
The other common fault on this driver board is the failure of the TL072 at the top corner of the board.
I would not expect this amp to function in any way that could be used for troubleshooting without half the drivers in the board.
The two 1D transistors in the center of the board run hot and fail often. Those get replaced, even if they check as OK with a meter.
The other common fault on this driver board is the failure of the TL072 at the top corner of the board.
I would not expect this amp to function in any way that could be used for troubleshooting without half the drivers in the board.
Attachments
Oh, sorry i will certainly do that, i was thinking the forum would not allow high resolution photos. The drivers were taken out to be tested but i did not have the time to put them back in. Thank you for the help, i will take out the driver board to check the listed components. My question is, should i be able to see the square wave on both sets of these fets? I just talked to the guy and he told me the amp would work but it did not have power, it must have had around 2 maybe 300W as he described. It kind of confused me that compared to ground i could see 63khz square wave on one set of the mosfets but on the others i could not. Also on the side where i could not see the square wave it would get hotter than the other side, as far as i know these amps should run cool unloaded, but after like 15 20 seconds i can barely hold my fingers on the warmer side of the amp. I am thinking maybe the fets are not from the same batch and that is why some are getting hotter..
Here are some pics of the gates of the fets. I have put back the driver trannys, checked for shorts on the driver board(the tl072s and the d1s) since i do not have too many smd parts on stock i did not replace them. Any ideas what should i check? Thanks again
Attachments
The amp appears to be working. Have you tried driving a signal into it to see if it will produce audio?
If you look closely under a bright light, you will see that the two 1D transistors have turned brown. If they haven't failed yet, they may soon. Even if the amp is now producing audio, replace them.
If you look closely under a bright light, you will see that the two 1D transistors have turned brown. If they haven't failed yet, they may soon. Even if the amp is now producing audio, replace them.
I know it will drive audio, because the guy was using it before bringing it to me, but what concerns me is that the fets are warming up, particularly on the side where the square wave looks well not square and the signal amplitude is a lot smaller. Fets are from the same batch but my idle current is around 2amps. Which seems kind of a lot to me in a D class? Correct me i might be wrong. I am basing this repair on some other repair guides. But they generally seem to be amps where you can leave the amp on for 20 minutes without cooling and the fets would be cool. Well here, umm after 20 seconds the low side is already untouchable so I will not leave them on for any lnger..
I will check this out, yes he is saying he went to a shop and bought exact replacements for every part. But as i know our local parts shops they are barbarians, so these must be fake STirf640s, they appear to be from the same batch but they look kinda fake too. Will check this out. Thanks again
The drive signals are referenced to the source leg of the FET. With a battery powered scope, you can measure directly across the gate-source legs. For someone using a mains powered scope, you have to use the ground reference (as it appears you did).
Buy from an authorized distributor when you buy FETs. Buying from ebay is more likely to get you counterfeit or second quality parts. Use octopart to find authorized distributors.
Buy from an authorized distributor when you buy FETs. Buying from ebay is more likely to get you counterfeit or second quality parts. Use octopart to find authorized distributors.
Ebay is betfer quality rhan our local shops, as far as i experienced. I have got countless desoldered/used parts sold as new.. it is okay when we are talking about something else but in an amp? Gotta buy parts from mouser or farnell, i do not seem to find other feputable sellers in my country. Thanks for the help, i will be back with pics to help future hifonics users
So here i attached some scope measurement of both the high and low sides (referenced to source this time) with IRF640s, i did chek and they need 640N s. I do not know if it is normal what i see, it looks kinda bad to me. I am putting some 640Ns in there and uploading some pics again.
Thank you!
Thank you!
Attachments
Okay so i have put the irf640Ns back in(original ones that did not get damaged when the amp burned out) and made some pictures. The fets are not heating up nearly as bad as before, they barely get lukewarm, so i am assuming this might be because they are old and stressed. Here are some pics of the switching action of each bank referenced to source.
I am going to order some 640Ns from a reputable dealer and i will update the post when i get them installed. Thanks again for the help!
I am going to order some 640Ns from a reputable dealer and i will update the post when i get them installed. Thanks again for the help!
Attachments
Right. Output filter inductors. Typical for virtually all class D amplifiers. They help block the high frequency carrier signal (typically about 80kHz.
.
http://www.ti.com/lit/an/slaa701a/slaa701a.pdf
Or google something like:
basic class D amplifier design output filter
Please start a new thread if you're not trying to help the OP with the repair.
.
http://www.ti.com/lit/an/slaa701a/slaa701a.pdf
Or google something like:
basic class D amplifier design output filter
Please start a new thread if you're not trying to help the OP with the repair.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.