Good morning everyone. I’m currently working on my first amplifier and made a rooky mistake and would like some advice before continuing. The amp is built around the LM3886 witch has a 84V max input rating. So I bought a 2x30V transformer thinking it would be 14V below the max voltage witch should be enough. But after measuring 80V over the power supply I realized it charges to the peak voltage and not the RMS voltages (duh). This left me with only 4V to spare, which made me a bit nervous but impatient as I was to test the amp I hooked up and it worked fine! Until a few days later I turn off my soldering iron near the amp which resulted a small bang from the speakers, the sound of sparks followed by silence 
.
I'm thinking of buying a new transformer with a 2x24V rating, which should give 67V. Would 17V below the max voltage rating be enough? is it wise to add some overvoltage protection to the amp’s PCB?
I was thinking about a 38V Zener with a resistor. Should be able to add it without to much Hassel. I want to avoid re designing and etching the PCB’s.
All advice is welcome, the more the better! or maybe some references.
(Getting a new transformer would be a good reason to start a guitar amp project with the old one, how annoying).
. I'm thinking of buying a new transformer with a 2x24V rating, which should give 67V. Would 17V below the max voltage rating be enough? is it wise to add some overvoltage protection to the amp’s PCB?
I was thinking about a 38V Zener with a resistor. Should be able to add it without to much Hassel. I want to avoid re designing and etching the PCB’s.
All advice is welcome, the more the better! or maybe some references.
(Getting a new transformer would be a good reason to start a guitar amp project with the old one, how annoying).
I assume that your soldering iron was a modern low-voltage one? If so, then it would have had transformer inside it. Therefore possibly being a reactive load, dumped some transients into your multi-plug that you might have had your new AMP circuit on while testing? That MIGHT have got magnified via your PSU arrangement?? Anyway, you might want to consider using appropriately specified MOV (Metal Oxide Varistor) components between your transformer and your bridge rectifier, so that it does not rectify the peaks coming on off the mains. If you have room for one of those filter units (the type with the integral IEC Socket) they are very handy for getting a cleaner AC going in. Using a MOV is very common (PC PSU units use them a lot, although the ones I have seen are VERY small, and can go short spectacularly). Use the highest Joule rating you can fit, and if possible an in-line fuse carrier for each MOV, to protect your PCB tracks from short events. ZENER: why not try an appropriate Voltage Regulator instead?
The chip not fail even at its maximum rated supply voltage. You need to be aware that the off load voltage of a transformer can be significantly higher than its rated value (which is at full load current). This will be quoted as the 'regulation factor' in the transformer data sheet.
A 22-0-22 VAC transformer is ideal for the LM3886.
A more likely explanation of the failure could be a burst of RF or arcing from the soldering iron which was then picked up and amplified causing the failure.
Microsecond and even millisecond transients on the mains supply are not going to impress onto a DC rail that has thousands of uF of reservoir capacitance.
A 22-0-22 VAC transformer is ideal for the LM3886.
A more likely explanation of the failure could be a burst of RF or arcing from the soldering iron which was then picked up and amplified causing the failure.
Microsecond and even millisecond transients on the mains supply are not going to impress onto a DC rail that has thousands of uF of reservoir capacitance.
To td47:
Ye the soldering iron converts to 18V but it was on a separate wall socket. I only need the 2 voltage rails so I think it’s a bit over the top to add a voltage regulator after the transformer. Funny you are mentioning the MOV just learned a bit about them yesterday. But they used them after the mains plug. Does it matter if you filter before or after the transformer? (Probably). I wanted to know if it’s advised to use a Zener as an extra safety, it would probably never conduct unless something goes wrong. Space is not a problem, still need to make the case and a friend of mine can get the material for free at the ship yard he works for. Not having a case wouldn’t have worked in my advantage either…
To mooly:
“A 22-0-22 VAC transformer is ideal for the LM3886” Thx!!
I would also have expected the capacitors to deal with the spikes, got 4 10.000uF witch should be more than enough, that’s why I thought it couldn’t hurt to use the amp with the high voltage PS (for the time being). But I also expected the lm3886 to clip instead of breaking with a high input signal.
After posting I saw that there are looooooooads of topics about the lm3886. I will do some more research tomorrow, after my exam…. thx for the help!
Ye the soldering iron converts to 18V but it was on a separate wall socket. I only need the 2 voltage rails so I think it’s a bit over the top to add a voltage regulator after the transformer. Funny you are mentioning the MOV just learned a bit about them yesterday. But they used them after the mains plug. Does it matter if you filter before or after the transformer? (Probably). I wanted to know if it’s advised to use a Zener as an extra safety, it would probably never conduct unless something goes wrong. Space is not a problem, still need to make the case and a friend of mine can get the material for free at the ship yard he works for. Not having a case wouldn’t have worked in my advantage either…
To mooly:
“A 22-0-22 VAC transformer is ideal for the LM3886” Thx!!
I would also have expected the capacitors to deal with the spikes, got 4 10.000uF witch should be more than enough, that’s why I thought it couldn’t hurt to use the amp with the high voltage PS (for the time being). But I also expected the lm3886 to clip instead of breaking with a high input signal.
After posting I saw that there are looooooooads of topics about the lm3886. I will do some more research tomorrow, after my exam…. thx for the help!
I'm not sure what you are thinking regarding zeners, but to use one as a clamp on an unregulated supply would be doomed to failure... there is just so much energy available at low impedance that the zener would just fail short circuit.
The soldering iron scenario is all a bit of guesswork, but don;;t underestimate just how 'loud' such RF transients can be. Its possible they could overload the front end of the chip causing some kind of latch up.
An RF filter to the input of any power amp is always a good idea.
The soldering iron scenario is all a bit of guesswork, but don;;t underestimate just how 'loud' such RF transients can be. Its possible they could overload the front end of the chip causing some kind of latch up.
An RF filter to the input of any power amp is always a good idea.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.