Ok, so a couple of amps died. I could do with some help
I have just had 2 class d amplifiers die on me and I do not understand why. I am using much less voltage than their max and they are supposedly really reliable little things. I was paralleling the amps with my 24V ~10A power supply. This is the product: https://www.ebay.co.uk/i/223709821051?chn=ps&norover=1&mkevt=1&mkrid=710-134428-41853-0&mkcid=2&itemid=223709821051&targetid=594652386360&device=c&mktype=pla&googleloc=1007011&poi=&campaignid=6466521308&mkgroupid=76196181743&rlsatarget=aud-629407027105:pla-594652386360&abcId=1140486&merchantid=109866676&gclid=CjwKCAiA5JnuBRA-EiwA-0ggPQIdW_wNci44yB2NbkFB9N2HsvjS2Ee_44WPe7w5SybirxT9QSFHRxoCqq0QAvD_BwE
Firstly the rear speakers' amp died and then I realised there wasn't much thermal paste on them so I put some on the final two, and now the one providing amp power to the centre and sub just went. What do I do, can I fix it? Can I return it? What did I do wrong? I am having a panic attack right now. No it isn't my preamp out puts, neither is it the wires, nor the speakers, nor the powersupply.
I heard, 0 distortion or clipping on the amps, there was no puff, no overheating, no popping when turning off, the only reason I realised is because I couldn't hear the speakers playing anymore
Also, the subwoofer and centre channel were 6ohms each not even 4 so I don't know why this happened
Also with this tiny amound of power the rear speakers were at 1/5 volume on the volume nob when it died. I don't even know if the boards are dead, I just don't seem to be getting an output from them anymore. The one for the front two stereo is still working but im not using it until i found out what happened. Also that link is for the same model with the same board but I didn't buy it from there, somewhere else.
Also I know the pos and neg wires are the right way round, anyway it was working for quite a while. Did I mention it wasn't heating up anyway.
all components look perfectly fine
The two amps that stopped working are detatched and the single amp seems to keep running fine.
I must have clipped it. It wasn't audible but I must have. It is the only explanation I can think of. All I need now is some help on what part of the circuit board went wrong so I can fix it. The reason I figured I must have clipped it is I found that clipping also occurs when an amp doesn't get enough current from the power supply. Derp: I gave each amp about 3A each and was asking much more. Weird I didn't acutally hear the clipping or it actually dying though (other than the speakers not producing sound)
p.s. this opening post jumps about a bit because they were seperate posts before I joines them. I forgot I could just edit. LOL
I would measure the DC voltage out of your power supply to see what it is actually reading. If also check the resistance (ohms) on your speakers again. Just to see if anything happened that you're not aware of. It sounds as if you had to high of voltage or too low impedance. I know you don't think that's an issue, but it is a quick check with a multimeter.
Quick check, you didn't somehow activate the MUTE function, did you?
Next, check the voltage ratings on the caps to make sure they are at least 50% over 36V. It is possible one or more died without you knowing. There are a number of ways to check if they are good, but start with a visual check for bulges or leaks.
Use a good lens to inspect the solder joints and traces. Cold or open joints should be visible upon inspection.
If nothing shows up, start by tracing the voltage from the power connections inward. Even though you probably don't have a schematic, most of these boards use something very close to what is on the data sheet for the chip, so have a look at that for reference.
Check the output chokes for continuity.
Come back after giving the amps (1 at a time with copious notes please) a good going over.
@123toid I realise it must have been some sort of clipping (even though it wasn't audible, because it happened when I turned the volume too high for the current supplied, unfortunately I thought I was making sure I wasn't breaking the boards by limiting the current available to them by putting them in parallel with the power supply, but of course the amps will only take however much current they need, I know that now because of that choice I must have broken something in the amp). Now that I have ordered some proper power supplies based on what we discussed would be the best wattage it should be fine. I was using my 3D printer's 24V meanwell power supply to test them, unfortunately I misunderstood that less available amps is worse not better. With these power supplies I wont be asking anywhere near as much of them as I just was. But I will check the output voltage from the printer's power supply just in case.
On what tvor said:
I didn't mute it, I noticed those soldering dots being marked mute and standby, I tested both with a screwdriver. (on the good board) Standby only lasted until it got in imput signal and mute only lasted until I lifted the screwdriver off). With the broken boards there was no change so it isn't that.
Both boards look great in terms of really thick tracks that are clean and unbroken. The soldering on both boards is maticulous. The decoupling capacitors are Sanyo (real) and are rated for 50V and are 470uF as I said I was only running 24V. I'll test if that was true later. Also I'll follow component paths testing for shorts or voltage contininuity later. I thought for now it's best to post my findings so far.
Thanks a lot guys, I hope I'll find more out later today.
I hope I can give back to the comminity like you have given to me. 😉
Just did the power supply test, I was getting 25.1V 10A with the amp connected I was still getting 25.1V. So that is done. 🙂
As I said it was probably giving the amps 24V 3.33A each that was pushing it too hard.
(too little amps for what I was asking as I already said). Does this make sense electronically?
I do hope it wasn't the chip!
Here's the Data Sheet for the chip they say they used. If you go over it, there is quite a bit of protection built in, most specifically to your point, under voltage and over current.
If you were using the same PS to power multiple units (you said you were), and the last ones continued to work while the others were still hooked up, I have to say that there isn't a short from the power rail to ground. If there was, it would have shut the PS down, if not temporarily, then more permanently.
I suspect there is an open circuit somewhere, maybe in one of the tiny passives on the board, or perhaps there was too much input voltage. It's very hard to say what really happened.
Looking at the photos in your link, I see there are some parts under the heatsink (not the TDA chip) that could be tested as well as the others outside there.
What does the switch labeled (ON 1 2) do?
Please know that I ask these questions as a means to rule out possibilities and narrow it down to hopefully fix what is wrong. It is all part of the troubleshooting process.
What does the switch labeled (ON 1 2) do? That increases or decreases the input gain.
Ah, okay. Didn't see any explanation in the description. Puts that to the side of the troubleshooting list / low suspicion.
My multimeter broke after trying to see how many amps were coming out of the power supply (clearly more than 5) so I need to get a new one before I test the boards (it was super cheap anyway). My third board that I have is working perfectly fine and really good. So either the other two were defects or I busted something. I didn't have the gain switches turned on as my motherboard surround outputs already send a line out signal (the upper voltage limit of the amplifier boards, I think). So next up is to check the components, and yes there are more components underneath the heat sink. I saw them when I replaced the thermal paste on the chip with some better stuff. They appear to be surface mounted components though (I don't know if they are replaceable or not). As I explained, I do not understand why only two of them went. The Philips surround speakers I was using to test it on were 3ohms (minimum) but I don't think that made a difference as the amp supplying power to the fronts (3ohms each) kept playing, the rear (3ohms each) broke and amp powering the centre+sub (6ohms each???) also broke. And yet most of the power was going to the FL and FR as I was listening to music and the sub was barely moving. So peculiar.
Someone did a review on one of the boards where he analysed some components. Maybe it will be useful?
Maybe I was quite unlucky and got a couple of boards with bad chips??? One of them that broke had a green QC Passed sticker though so I don't know.
While you are waiting for your new meter, there's a way to check the capacitors without a meter, at least to see if they are intact and functional.
Grab cheap amp, Iike a LM386 type. Hook up a relatively full range speaker to that amp. Now, this is the test area part: in the input line between the source and amp, cut the "positive" lead and hook up a probe lead to each side of the cut.
To test the cap, put the probes on either side and play some music. If the cap is functional, it will let the audio through, and the value will determine the frequencies it will pass.
Music = functioning cap
No Music = bad cap
Now, this won't verify the cap's leakage, ESR, or any other parameter, but it will help eliminate direct faults.
Thanks to shango066 for that tip.
Just to clarify:
Your set up is basically a 5.1 system - 2 front, 2 rear, a center and a sub, powered with 3 of these stereo amps. 6 speakers total, each driven by 1 channel of an amp (6 channels total).
The power is from a single "power brick", with a rating of 240 watts (24V x 10A).
Do I have this correct?
(verifying for a friend - 😀 )
@tvor-ceasar Yes that is the setup I had for testing, I had bought 3 smaller meanwells (1 for each power supply so I could get more output anyway) which is good because it turns out not to be a good mix when different amps are providing different loads when in parallel with 1 SMPS. All good now though as they work when on their own supply. (Explained below). If your buddy wants to repeat my setup I am willing to help in any way I can. 🙂
You wouldn't believe it. The protection circuits had kicked in when I wasn't providing enough current to the amps that needed it, I tested them again yesterday (I thought why not, as they had been off for a few days), but there must have been some residual power in the capacitors or something and why the protection circuits didn't deactivate. All is good now though as they both work. I am over the moon. 🙂
People be aware: do not ask the amp of too much when you aren't giving it the full amount of amps it needs, if it doesn't have protection circuits it will break!
Topic saved as solved. 🙂
I am now hoping to log my full speaker design process as I want to share it with everyone.
123Toid, if you could, could you delete the Dolby Atmos Bookshelf speaker design post as I want to restart it but lay it out better