I guess that it had to happen. Such a small board and such big fingers....
This afternoon I tried to fire up my test version of the SI amp. I have another working version luckily. As I went to connect it to my transformer volume control (TVA) I had a moment of lucidity and decided to check for DC on the input. Lo and behold I had DC on one channel fluctuating oddly up to nearly half a volt. Since there is a blocking cap on the input to keep voltage from going back upstream I figured that I had either a problem amp or else I had misoldered something. The only mods I had made were to replace the output chokes and to wire up the power a bit differently by going directly to the terminals of the big supply cap. The input connections were to the same locations but without the original board mounted connector which I had removed.
Any ideas? The other channel input measures zero volts as do both channels of my working amp.
Thanks a lot,
Paul
This afternoon I tried to fire up my test version of the SI amp. I have another working version luckily. As I went to connect it to my transformer volume control (TVA) I had a moment of lucidity and decided to check for DC on the input. Lo and behold I had DC on one channel fluctuating oddly up to nearly half a volt. Since there is a blocking cap on the input to keep voltage from going back upstream I figured that I had either a problem amp or else I had misoldered something. The only mods I had made were to replace the output chokes and to wire up the power a bit differently by going directly to the terminals of the big supply cap. The input connections were to the same locations but without the original board mounted connector which I had removed.
Any ideas? The other channel input measures zero volts as do both channels of my working amp.
Thanks a lot,
Paul
Re: Re: SI Mod Problem
I took a large pair of wire cutters and cut the plastic strip and pins in half. that way you can just heat the pins from the bottom and gently push them up when the solder melts (certain soldering iron tips work better for this than others). than you can grab the hacked end of the pin with a pair of needle nose pliers, apply a little heat to the pin, and pull it out.
Panelhead said:
I took a large pair of wire cutters and cut the plastic strip and pins in half. that way you can just heat the pins from the bottom and gently push them up when the solder melts (certain soldering iron tips work better for this than others). than you can grab the hacked end of the pin with a pair of needle nose pliers, apply a little heat to the pin, and pull it out.
Thanks for the speedy replies.
I looked really carefully for a solder bridge. That was my first guess. I scraped between the tracings and cleaned up with some isopropyl alcohol. The wires are in the right holes. I am going to give it another look tomorrow. It must be a solder bridge or broken tracing somewhere.
I cut off the adaptor using the method decribed by BWRX. It worked fine on my first amp but it looks like I might have messing up on this one. I sure wish someone would make a nice DIY version of this board where the signal path devices were through-hole and there was more room to play
I want to replace the coupling caps with some off-board polypropylenes but have not gotten the courage up to unsolder those miniscule smd caps yet...
Best,
Paul
I looked really carefully for a solder bridge. That was my first guess. I scraped between the tracings and cleaned up with some isopropyl alcohol. The wires are in the right holes. I am going to give it another look tomorrow. It must be a solder bridge or broken tracing somewhere.
I cut off the adaptor using the method decribed by BWRX. It worked fine on my first amp but it looks like I might have messing up on this one. I sure wish someone would make a nice DIY version of this board where the signal path devices were through-hole and there was more room to play
I want to replace the coupling caps with some off-board polypropylenes but have not gotten the courage up to unsolder those miniscule smd caps yet...
Best,
Paul
Since the soldering of the rest of the board went well I decided to go ahead and bypass the input stage with an offboard 2.2mf poly and a 22k metal film resistor. Hopefully I will also bypass the problem
On another note. I replaced the output chokes on my "good" amp while I was moping over the experimental one. Boy those solder pads come off easily. I had to do some serious rerouting and jury rigging by the time I was done. I now have both outputs wired directly to one lead of the new chokes. If you do this be SURE to choose the correct end. On my amp the 5 ohm resistors I put on my outputs for headphone use got VERY hot when I chose the wrong end of the choke to solder to. I bet it would fry a speaker. I now use a cheapo PC speaker as a test speaker. On my experimental amp I have the power going directly to the pins of the big power cap. Works well.
The good amp does sound better with the new inductors. It seems bit more open and relaxed. This thing just keeps getting better...
BTW. Do any of you fuse the battery?
Best,
Paul
On another note. I replaced the output chokes on my "good" amp while I was moping over the experimental one. Boy those solder pads come off easily. I had to do some serious rerouting and jury rigging by the time I was done. I now have both outputs wired directly to one lead of the new chokes. If you do this be SURE to choose the correct end. On my amp the 5 ohm resistors I put on my outputs for headphone use got VERY hot when I chose the wrong end of the choke to solder to. I bet it would fry a speaker. I now use a cheapo PC speaker as a test speaker. On my experimental amp I have the power going directly to the pins of the big power cap. Works well.
The good amp does sound better with the new inductors. It seems bit more open and relaxed. This thing just keeps getting better...
BTW. Do any of you fuse the battery?
Best,
Paul
Well, I learned some more this evening. I decided to go ahead and bypass the stock input circuit with a 2.2mf and a 22k resistor since it had an issue anyway. After doing this I now had correct voltages on either side of the caps. 0v on input side and 2.3v bias on the amp side. Unfortunately one channel was dead. Turned out that I had miswired one pair of outputs: wrong holes. 
I had to take out the board so that I could get to the underside. In doing this I pulled off the tiny new input wires. Bummer. They were much harder the second time around and I ended up losing a pad. I found alternate attachment points at one end of the feedback resistors R4,5. Unfortunately only one channel came up after this debacle. Basically this board is toast. I am now batting 50%. The good news is that the "live" amp sounds so good that I will just pick up and try again. I really want to see what those input mods will do. I should have another SI from Parts Express coming in a couple of days.
I sure hope that someone makes a really stout DIY board soon with big ol' tracings and through hole parts in the signal path... and smd parts presoldered!
Here is what I learned.
1. You need to plan out your mods and REALLY limit the number of times you take the board in and out. Those pads are so weak that even wire torque can pull them up. They cannot take much unsoldering or resoldering.
2. Use extremely small wires with very thin insulation to wire the offboard input cap and resistor to the vacated pads. Use glue or a hot glue gun to tack the wire to the board. You may only get one shot at this because the pads are so flimsy. You can also use the glue gun to tack down the big cap if necessary.
Now Listening to Dave Brubeck on my live modded SI with some AKG K1000's...
Best,
Paul
I had to take out the board so that I could get to the underside. In doing this I pulled off the tiny new input wires. Bummer. They were much harder the second time around and I ended up losing a pad. I found alternate attachment points at one end of the feedback resistors R4,5. Unfortunately only one channel came up after this debacle. Basically this board is toast. I am now batting 50%. The good news is that the "live" amp sounds so good that I will just pick up and try again. I really want to see what those input mods will do. I should have another SI from Parts Express coming in a couple of days.
I sure hope that someone makes a really stout DIY board soon with big ol' tracings and through hole parts in the signal path... and smd parts presoldered!
Here is what I learned.
1. You need to plan out your mods and REALLY limit the number of times you take the board in and out. Those pads are so weak that even wire torque can pull them up. They cannot take much unsoldering or resoldering.
2. Use extremely small wires with very thin insulation to wire the offboard input cap and resistor to the vacated pads. Use glue or a hot glue gun to tack the wire to the board. You may only get one shot at this because the pads are so flimsy. You can also use the glue gun to tack down the big cap if necessary.
Now Listening to Dave Brubeck on my live modded SI with some AKG K1000's...
Best,
Paul
Off board caps
I am with you. I pull my apart too much. This is why I have lived with the crappy 2.2 ufd ceramics.
I hope to build some with 2.2 ufd film caps in the near future. But for now the ceramics are what it is.
The pad for the input coupling caps are just so tiny. Guess the method of shorting the inductors and coupling, removing the shunt resistiors and caps is best. Then you can you the input holes to mount to.
George
I am with you. I pull my apart too much. This is why I have lived with the crappy 2.2 ufd ceramics.
I hope to build some with 2.2 ufd film caps in the near future. But for now the ceramics are what it is.
The pad for the input coupling caps are just so tiny. Guess the method of shorting the inductors and coupling, removing the shunt resistiors and caps is best. Then you can you the input holes to mount to.
George
Re: Off board caps
I agree here as well. However, I'm going to be honest -- don't waste your time stripping and doing the input mods off board. I really don't think that the improvement is worth it, nor is the pain. I'm pretty convinced that there's 3 ways to do it:
1) Go with the lower value black gates. This is easy, since they fit on board with little problem.
2) Run some small jumpers via fine wires to your film caps (maybe some 63v MKP's or something)
3) Go for some 1uF or 3uF boxed stacked film caps. Those, too, will probably fit on the board, similar to the blackgates.
Based on my own experience of toasting one board, and not hearing that much of an improvement on the other (i.e. after yanking off all the highpass stuff, etc.), I think it's better to just leave that stuff on with the SI's.
Panelhead said:
I agree here as well. However, I'm going to be honest -- don't waste your time stripping and doing the input mods off board. I really don't think that the improvement is worth it, nor is the pain. I'm pretty convinced that there's 3 ways to do it:
1) Go with the lower value black gates. This is easy, since they fit on board with little problem.
2) Run some small jumpers via fine wires to your film caps (maybe some 63v MKP's or something)
3) Go for some 1uF or 3uF boxed stacked film caps. Those, too, will probably fit on the board, similar to the blackgates.
Based on my own experience of toasting one board, and not hearing that much of an improvement on the other (i.e. after yanking off all the highpass stuff, etc.), I think it's better to just leave that stuff on with the SI's.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.