Hi,
Today i powered up my LM3886 amp board as i have done before but this time thing went wrong. Its powering a subwoofer but this combination of psu, amp board, imput and sub have been used before with sucess.
today the sub is in a new box i built myself. i powered it up without any input and everything as well so shut it down and turned it back on with music input from my iphone (as have doen before) but this time things went wrong.
after about 1/2 a second smoke and sparks fly from an amp chip. the board being used is: (paralelle btl configuration)
300W Mono LM3886 X 4 + NE5532 POWER AMPLIFIER KIT | eBay
The chip that blew is the 3rd from the left (in the main pic) and pin 4 (what i believe to be V-)of this chip is completely burned out.
I checked the psu afterward and it is working fine giving the +- 35 V it should.
If anyone can help me or have any ideas it would be much appreciated.
the rest of the board looks fine so i can solder a new chip in but will it just happen again is the question?
thanks
Dion
Today i powered up my LM3886 amp board as i have done before but this time thing went wrong. Its powering a subwoofer but this combination of psu, amp board, imput and sub have been used before with sucess.
today the sub is in a new box i built myself. i powered it up without any input and everything as well so shut it down and turned it back on with music input from my iphone (as have doen before) but this time things went wrong.
after about 1/2 a second smoke and sparks fly from an amp chip. the board being used is: (paralelle btl configuration)
300W Mono LM3886 X 4 + NE5532 POWER AMPLIFIER KIT | eBay
The chip that blew is the 3rd from the left (in the main pic) and pin 4 (what i believe to be V-)of this chip is completely burned out.
I checked the psu afterward and it is working fine giving the +- 35 V it should.
If anyone can help me or have any ideas it would be much appreciated.
the rest of the board looks fine so i can solder a new chip in but will it just happen again is the question?
thanks
Dion
probably !
We get more queries on problem multi-chipamp assemblies than on almost any other chipamp topic.
We get more queries on problem multi-chipamp assemblies than on almost any other chipamp topic.
thanks haha but no ideas why? i mean its worked before is exactly the same setup except a sub box change..
without a schematic and a detailed closeup of both sides of the unpopulated PCB, we have no hope of easily helping you to sort your broken amplifier.
Go back to your supplier and ask for customer service.
Go back to your supplier and ask for customer service.
which makes it a bpa200. I wonder where 300W comes from?
yes.
If one takes the risk of buying from China direct, then one must also take the risk that if problems become evident then the buyer is almost alone. Paypal and Ebay may be able to refund money in event of fraud, but kits that work and then fail are outside what they and us can sort.
If one takes the risk of buying from China direct, then one must also take the risk that if problems become evident then the buyer is almost alone. Paypal and Ebay may be able to refund money in event of fraud, but kits that work and then fail are outside what they and us can sort.
from those you can reverse engineer the schematic.
Then add on device names/types.
Finally determine normal operation and possibly guess at what is wrong with the design and/or assembly.
save this broken amp until you have learned how it should operate and can debug it your self.
In the meantime buy/build what you can debug your self or with the help of customer service (if it exists).
In the meantime buy/build what you can debug your self or with the help of customer service (if it exists).
asked for a copy of the schematic. we'll see how lucky i am haha
there isnt really anything i can debug myself. kinda just hoped it workes the way it should haha
there isnt really anything i can debug myself. kinda just hoped it workes the way it should haha
I don't see any decoupling caps at or near any of the LM3886. Only the electrolytes but a ~100nF ceramic at or close to the power pins is desired.
He doesnt have a schematic from his supplier but hes sending me a new chip to solder in. I'll have to do that then run it from a proper supply and input in a lab. (it wont be damaged with no speaker attached will it?). I'll repace the op amp as well with another i have.
as for the decoupling caps, how could i add those in?
Dion
as for the decoupling caps, how could i add those in?
Dion
i just remeasure the voltage input and it was +- 34.8 dc. the max for the 12V regulators on board is 35V so could there have been a spike and damaged these at a power on? i worked at a transformer place last year (and know them quite well so they will modify my trannie for free) so i might take it too them and get it taken down to 23V AC (ie 31V DC).
also as it was the V- line that blew, should i replace the -12V regulator or that shouldnt matter (as i expect the V- going in was the full -35V)
Dion
also as it was the V- line that blew, should i replace the -12V regulator or that shouldnt matter (as i expect the V- going in was the full -35V)
Dion
DionG,
I had a very similar problem with a LM3886 x 3 build. Thanks to Andrew's input I was able to trace down the problem with a dim bulb tester. After the green smoke and sparks I replaced almost everything and especially the on board caps. Two of the four actually spit out oil and were convex after the fireworks. My board uses many surface mount resistors and caps and I found atleast one suspected solder bridge on a SDM resistor.
After the rebuilds I isolated each leg of amp by removing the 0.1 resistors and using the bulb tester one segment at a time. I found two of the three LM3886s had also been damaged. Removing the LMs with first pass using desoldering braid and then a heat gun was not too difficult.
A valuable tool was an inexpensive laser thermometer ($29) to check cap tops as the amp was powered on. They stayed at room temps.
My board also has pots to adjust dc offset and those adjustments were again done one segment at a time.
In my case I can't blame the vendor as I purchased boards only and bought the parts from U.S. sources. Even so, Stanton Tin from Jim's Audio on eBay made himself available at every step of the rebuild.
I don't know if this is helpful but I learned to never power a new amp build without the use of a dim bulb tester. You can Google it for more info.
Good Luck.
I had a very similar problem with a LM3886 x 3 build. Thanks to Andrew's input I was able to trace down the problem with a dim bulb tester. After the green smoke and sparks I replaced almost everything and especially the on board caps. Two of the four actually spit out oil and were convex after the fireworks. My board uses many surface mount resistors and caps and I found atleast one suspected solder bridge on a SDM resistor.
After the rebuilds I isolated each leg of amp by removing the 0.1 resistors and using the bulb tester one segment at a time. I found two of the three LM3886s had also been damaged. Removing the LMs with first pass using desoldering braid and then a heat gun was not too difficult.
A valuable tool was an inexpensive laser thermometer ($29) to check cap tops as the amp was powered on. They stayed at room temps.
My board also has pots to adjust dc offset and those adjustments were again done one segment at a time.
In my case I can't blame the vendor as I purchased boards only and bought the parts from U.S. sources. Even so, Stanton Tin from Jim's Audio on eBay made himself available at every step of the rebuild.
I don't know if this is helpful but I learned to never power a new amp build without the use of a dim bulb tester. You can Google it for more info.
Good Luck.
Last edited:
DionG. Steep learning curve isn't it!😉
I built several LM38** amps some years ago. The best way to understand how these things work is to build one point to point i.e without a circut board. as you are joining all the required parts together, you know what should go where and why. If it goes wrong you have a much better idea of why.
I used this information for both 3886 and 3875 amps. I also built a bridge 3886by using DD's schematic and adapting the schematic from the Nat Semiconductor spec sheet. I thought I was really clever. Looked like a rats nest, but at least it worked.
Have fun.
Building a Gainclone chip amp
Building a Gainclone chip amp with the LM3886 chip.
I built several LM38** amps some years ago. The best way to understand how these things work is to build one point to point i.e without a circut board. as you are joining all the required parts together, you know what should go where and why. If it goes wrong you have a much better idea of why.
I used this information for both 3886 and 3875 amps. I also built a bridge 3886by using DD's schematic and adapting the schematic from the Nat Semiconductor spec sheet. I thought I was really clever. Looked like a rats nest, but at least it worked.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Have fun.
Building a Gainclone chip amp
Building a Gainclone chip amp with the LM3886 chip.
- Status
- Not open for further replies.