I'm working on a friend's subwoofer amp and I'm worried about one of the eight-pin DIP IC locations. I'll skip over the history to get to the important part. This one IC location has gotten badly chewed up - there's a chance someone was in there before me and worked on the same IC. By this point some of the pads on the non-component side are missing; I'm not so worried about missing pads where there is no trace on that side but there are some pads missing where there are traces leading away. There is at least one layer of traces buried within the board. I intend to put an IC socket there like I have for all the other ICs but I'm afraid that if I go ahead and solder a socket into place, I will be left with some open connections and I'd never be able to troubleshoot it especially since I have no schematic and apparently the universe has made it unobtainable.
So I have an idea to proceed forward with and I'd like to know if any of you have a better one.
It turns out that I have some 3/64" brass tubing that was obtained from a hobby shop. I thought what I might do is drill the PC board holes out with a 3/64" bit, scrape away the green coating from the traces I can see on both sides, tin the outside of a length of the tubing, cut the tubing into roughly 1/8" lengths, insert those lengths into the holes I drilled out, and heat them up with an iron, also soldering the tubing to the traces where they exist. I've already verified that the IC socket pins will fit within the I.D. of the tubing without difficulty. It may be a bit tricky to solder the IC socket to the pieces of tubing; perhaps the socket's pins have to be held partway out of the tubing for that to work. My hope is that when heated, the tinning on the outside of the tubing will grab any embedded traces.
Does anyone have any thoughts on how I should involve flux? I have both liquid rosin and paste flux.
So I have an idea to proceed forward with and I'd like to know if any of you have a better one.
It turns out that I have some 3/64" brass tubing that was obtained from a hobby shop. I thought what I might do is drill the PC board holes out with a 3/64" bit, scrape away the green coating from the traces I can see on both sides, tin the outside of a length of the tubing, cut the tubing into roughly 1/8" lengths, insert those lengths into the holes I drilled out, and heat them up with an iron, also soldering the tubing to the traces where they exist. I've already verified that the IC socket pins will fit within the I.D. of the tubing without difficulty. It may be a bit tricky to solder the IC socket to the pieces of tubing; perhaps the socket's pins have to be held partway out of the tubing for that to work. My hope is that when heated, the tinning on the outside of the tubing will grab any embedded traces.
Does anyone have any thoughts on how I should involve flux? I have both liquid rosin and paste flux.
Boards have an even number of layers, so this is probably 4 layer or maybe 6. Do not drill out any holes, they are plated and you'll destroy them and lose continuity with inner layers probably... Or worse short to layers you shouldn't be connected to...
Do you have a photo of the damage so we can see how bad?
Do you have a photo of the damage so we can see how bad?
I would go ahead and solder in the new socket regardless. Then do some resistance tests to ensure every pin is connected where it should be. If you find one open circuit, then just use an external wire link to hook it up. Would be a bit easier if there's another channel to compare against, otherwise you're going to have to do some reverse engineering.
Last edited:
I don't have a good way to take macro photos.
I realize the holes are (were?) plated on the inside but at this point I don't know how much continuity has already been lost and won't be restored by just soldering a socket in. The point of using the brass tubing that's tinned on the outside is to reestablish continuity with every trace coming to each hole. You have a point about hitting traces that shouldn't be in contact; I will look and see if there's any place on either visible side of the board if there are any traces anywhere that thread between 0.1" hole spacing.
@kaptiaudio Without a schematic I'd be hard pressed to distinguish an open fault from not supposed to be connected in the first place.
As @Mark Tillotson says DO NOT drill out the plated through holes. You will lose connection with the sub layers. If only the lower layer has pads missing you can buzz out connection to the trace. If missing scapre the resist from the trace and use a fine wire to link back to the chip pins. You may have a problem removing the suspect chip, best to clip the pins and remove one at a time.
Scrap the board, or get the schematic, stick socket in place, use fine insulated wires (motor winding wires) to connect to the rest of the circuit. Join them to different places in the circuit.
Had this problem on a HP PC motherboard.
Two repairs by a tech failed, so bought another, quite satisfactory result, it was like $25, works better than the old one.
Do this only when you cannot find a replacement.
Had this problem on a HP PC motherboard.
Two repairs by a tech failed, so bought another, quite satisfactory result, it was like $25, works better than the old one.
Do this only when you cannot find a replacement.
There are no replacement boards. All attempts to find a schematic has failed. The manufacturer was bought up many years ago and the current trademark holder has no information.
I went forward with soldering the IC socket in the problem location without doing the drilling and tinned brass tubing business. I tried shining a flashlight through the board and best I can tell, there are not traces within it and furthermore, there are no traces leading away from the pads on the component side of the board. So I went one socket pin at a time knowing by inspection that the noncomponent side had socket pins with either nothing to solder to at all or with a trace that was broken near where the pad was. One downside was that with the socket in place I could not see the pad on the component side but I reckoned that there is likely no harm done if the solder doesn't make it all the way through the hole to the component side. I soldered jumper wires between to socket pins and components on traces that weren't making contact with pins anymore.
I populated the sockets with op amp ICs and as I suspected, the board is a very steep lowpass filter that isn't letting through much of anything above 50Hz. The actual amplifier is on another board and whereas I am getting speaker output I don't know if everything is working right yet.
I went forward with soldering the IC socket in the problem location without doing the drilling and tinned brass tubing business. I tried shining a flashlight through the board and best I can tell, there are not traces within it and furthermore, there are no traces leading away from the pads on the component side of the board. So I went one socket pin at a time knowing by inspection that the noncomponent side had socket pins with either nothing to solder to at all or with a trace that was broken near where the pad was. One downside was that with the socket in place I could not see the pad on the component side but I reckoned that there is likely no harm done if the solder doesn't make it all the way through the hole to the component side. I soldered jumper wires between to socket pins and components on traces that weren't making contact with pins anymore.
I populated the sockets with op amp ICs and as I suspected, the board is a very steep lowpass filter that isn't letting through much of anything above 50Hz. The actual amplifier is on another board and whereas I am getting speaker output I don't know if everything is working right yet.
Four layer PCBs are rare in consumer audio with through hole ICs due to their high costs. A subwoofer is hardly size constrained.
Here the 2.1 kits use a 4558 to filter the subwoofer sound and drive the subwoofer chip amp, so it should be easy to build a new circuit from current parts.
Or just buy a ready plate amp .
2.1 means stereo + woofer, 4.1 is 4 channels + woofer, and 5.1 / 7.1 plate amps are also sold here, a 2.1 plate amp with 3 D2030 chips, 4558, 7805 for FM module, etc. starts at $2.50 retail. Just add 12-0-12 supply transformer, and pots for bass, treble, volume....ready in half an hour.
Higher powered versions are also available, and you seem to need only the pre amp section, not the main amp, right?
There is plenty of information on the web about building a crossover with known frequency response, use that to get your required curve.
This will be easier and more robust than looking for an equivalent board, it will give the same result.
Or just buy a ready plate amp .
2.1 means stereo + woofer, 4.1 is 4 channels + woofer, and 5.1 / 7.1 plate amps are also sold here, a 2.1 plate amp with 3 D2030 chips, 4558, 7805 for FM module, etc. starts at $2.50 retail. Just add 12-0-12 supply transformer, and pots for bass, treble, volume....ready in half an hour.
Higher powered versions are also available, and you seem to need only the pre amp section, not the main amp, right?
There is plenty of information on the web about building a crossover with known frequency response, use that to get your required curve.
This will be easier and more robust than looking for an equivalent board, it will give the same result.
No schematic, so a little handicapped because of it.
OP has not given IC number, just 8 pin DIP...seems a new to hobby person.
OP has not given IC number, just 8 pin DIP...seems a new to hobby person.
@NareshBrd TL072. That's a dual op amp, if you didn't know 😉. As mine was a physical PC board question, I didn't find the IC number relevant but I guess trolls gotta troll. Incidentally, "new to 'hobby' person" got the amplifier working last night. With no schematics. After having diagnosed and replaced one of the four power transistors, a bad filter cap in one of the three separate +-15VDC supplies, a failed optoisolator, and two voltage regulators. I'm not exactly new nor unschooled (I was offered a job with one of the biggest audio repair shops in the Southeast on the basis of my experience and the "entrance exam" they had me take but I had to turn it down) but I haven't seen everything and this amplifier is way outside anything I've encountered.
Oh, BTW @NareshBrd, I suppose ripping out and replacing all of the electronics (to include adapting a COTS plate amp; this thing is already a plate amp, point of fact) could be an option but this is just one side; the other speaker with its identical amp is 800 miles away in storage so I'd wind up having to replace both amps to make them the same. I'd much rather fix the one on my bench.
Oh, BTW @NareshBrd, I suppose ripping out and replacing all of the electronics (to include adapting a COTS plate amp; this thing is already a plate amp, point of fact) could be an option but this is just one side; the other speaker with its identical amp is 800 miles away in storage so I'd wind up having to replace both amps to make them the same. I'd much rather fix the one on my bench.
It's working. While scrapping is always an option, this amp is out of a pair of tower speakers that retailed for $2800 new. The speakers are unpowered except for the sub and they use the incoming speaker connection as the input to the subwoofer amp. So it's all kind of a package deal. The owner has the option of running these things with the sub defeated and adding a standalone powered subwoofer to his setup, I guess.
Read my post again...4558 is also a dual op amp, and used in a low frequency filter circuit, which is what you needed.
I offered a solution to your problem.
And it would have worked as a filter.
Many Americans are wise guys, they think (Germans too), that they are the only people with intelligence.
Trolling means teasing, learn some wisdom, it will help you in life.
I offered a solution to your problem.
And it would have worked as a filter.
Many Americans are wise guys, they think (Germans too), that they are the only people with intelligence.
Trolling means teasing, learn some wisdom, it will help you in life.