I am trying to find the problem. And the left channel was so easy (or I was lucky...) that I thought I was in the clear with my IGC.
I resoldered one suspect joint (V++ power in at the 1000uf cap) which looked promising, but... nuthin. I was too pressed for time to follow the whole circuit through with the ohmmeter but plan to do that late this evening.
Any help on where to concentrate would be much appreciated.
Thanks to Nuuk and the DecDun pages for the hard-to-miss warnings to check the offset before hooking up speakers!
GnD
I resoldered one suspect joint (V++ power in at the 1000uf cap) which looked promising, but... nuthin. I was too pressed for time to follow the whole circuit through with the ohmmeter but plan to do that late this evening.
Any help on where to concentrate would be much appreciated.
Thanks to Nuuk and the DecDun pages for the hard-to-miss warnings to check the offset before hooking up speakers!
GnD
That's a hard one to call Graham. I assume that you have checked thatyou have the correct value resistors in the right places!
Anyway, good luck with that multimeter. Do things slowly and logically and I am sure you will find the problem.
Anyway, good luck with that multimeter. Do things slowly and logically and I am sure you will find the problem.
Yeah, I spent so much time reading and studying the info (only occasionally get to set up my temporary workbench and get my hands dirty with kids and full-time+ job) that I had the thing memorized long ago.
When I go thru it and find out where my voltages start getting screwy, I'll know enough to ask questions. A quick check shows that the rail voltages (comparing right and left) do not come close to matching, (right rail voltage is way too low, along with high DC offset on the right output) and left rail voltages are both nominal for the secondaries I have.
At least that much makes sense to me. I have to find out where the right channel gets weird and maybe that will get me in the right neighborhood to fix it.
Thanks again for your great site. I hope to do you proud with my enclosure, assuming I can fix my right channel problems. Heck, maybe I will go deaf in my right ear and I won't have to worry about it!
🙂
GnD
When I go thru it and find out where my voltages start getting screwy, I'll know enough to ask questions. A quick check shows that the rail voltages (comparing right and left) do not come close to matching, (right rail voltage is way too low, along with high DC offset on the right output) and left rail voltages are both nominal for the secondaries I have.
At least that much makes sense to me. I have to find out where the right channel gets weird and maybe that will get me in the right neighborhood to fix it.
Thanks again for your great site. I hope to do you proud with my enclosure, assuming I can fix my right channel problems. Heck, maybe I will go deaf in my right ear and I won't have to worry about it!
🙂
GnD
Are you sharing rectifiers, or does each rail have seperate bridge rectifiers? If so, start checking there.
The obvious point to start is to check the DC delta at the inputs. If it is there, then multiply by the gain. If that is your offset, it's not the amp. If not, the chip may be at fault.
If the offset at the feedback point into the input is different from the offset at the output, you want to check the elcap in the feedback network. (Does this thing have 100% DC feedback?).
Let us know what you measure. Much more intelligent than doggedly starting to measure all R's with an ohmmeter.
Jan Didden
Edit: A chip with 4mV offset in a closed loop gain of some 45 would give 180mV if you don't have 100% DC feedback, and be fully within spec.
If the offset at the feedback point into the input is different from the offset at the output, you want to check the elcap in the feedback network. (Does this thing have 100% DC feedback?).
Let us know what you measure. Much more intelligent than doggedly starting to measure all R's with an ohmmeter.
Jan Didden
Edit: A chip with 4mV offset in a closed loop gain of some 45 would give 180mV if you don't have 100% DC feedback, and be fully within spec.
Nuuk-
I found my problem, and it was the same as yours, when you built your first GC and had a faulty umbilical cord. I have the GC in board version now, and as such, was using crimped spades for connectors from the bridges to the rails.
A spade connector where I was splitting my V++ into right and left channel had an intermittant connection for the right channel V++. When I tested it earlier, the pressure from the probe cuased it to be temporarily fine.
Everything is fine, they amp sounds clean as hell (can't judge yet, burning in with el cheapo Yamaha speakers) and absolutely zero noise. That's good, considering my board amp version left a lot of wires long for enclosure flexibility and trimming later. The whole damn thing is practially an antenna farm!
Thanks again for your comprehensive site and endless advice. If you ever are traveling through the States, you have a place to crash 20 minutes from Washington, DC.
My next project is just a re-enclosing bit on a crummy-looking (but great-sounding) META 42 headphone amp I bought. Putting it in an beautiful wooden cigar box.
After that, I will enclose the IGC (have to decide on how many inputs and such, plus some other stuff) and then I am looking to venture over to another area of DIYAudio and build some single-driver Fostex loudspeakers.
Will post pix as I work my way thru the IGC enclosure. Peter as been very inspiring in that area. Dunno if I am gonna be able to do something as unorthodoxed as your tube enclosures - I have to design in a high WAF, you know. 🙄 (Happy wife, happy life...) I am thinking black thick aluminum front/back/bottom chassis with teak sides and top...
Thanks again for all of your help to date and I am serious about the place to crash!
GND
I found my problem, and it was the same as yours, when you built your first GC and had a faulty umbilical cord. I have the GC in board version now, and as such, was using crimped spades for connectors from the bridges to the rails.
A spade connector where I was splitting my V++ into right and left channel had an intermittant connection for the right channel V++. When I tested it earlier, the pressure from the probe cuased it to be temporarily fine.
Everything is fine, they amp sounds clean as hell (can't judge yet, burning in with el cheapo Yamaha speakers) and absolutely zero noise. That's good, considering my board amp version left a lot of wires long for enclosure flexibility and trimming later. The whole damn thing is practially an antenna farm!
Thanks again for your comprehensive site and endless advice. If you ever are traveling through the States, you have a place to crash 20 minutes from Washington, DC.
My next project is just a re-enclosing bit on a crummy-looking (but great-sounding) META 42 headphone amp I bought. Putting it in an beautiful wooden cigar box.
After that, I will enclose the IGC (have to decide on how many inputs and such, plus some other stuff) and then I am looking to venture over to another area of DIYAudio and build some single-driver Fostex loudspeakers.
Will post pix as I work my way thru the IGC enclosure. Peter as been very inspiring in that area. Dunno if I am gonna be able to do something as unorthodoxed as your tube enclosures - I have to design in a high WAF, you know. 🙄 (Happy wife, happy life...) I am thinking black thick aluminum front/back/bottom chassis with teak sides and top...
Thanks again for all of your help to date and I am serious about the place to crash!
GND
That's great news Graham.
Te next best thing to building a Gainclone and hearing what comes out of it is to help others do the same.
The fact that the FAQ page seems to be helping in that respect justifies the time spent in creating it.
And thanks for the offer of a place to crash. I did tour the USA in 1981 but who knows, I may get back again one day. I'll know where to come by listening for the music as I have a hunch the sound made with some Fostex loudspeakers will be just what I like! 😉
And for any newbies out there, take note of Graham's experience. Often pushing the probe of a multimeter onto a connection will make it OK while the probe is there giving a false impression that the connection is OK when it is not.
Te next best thing to building a Gainclone and hearing what comes out of it is to help others do the same.
The fact that the FAQ page seems to be helping in that respect justifies the time spent in creating it.
And thanks for the offer of a place to crash. I did tour the USA in 1981 but who knows, I may get back again one day. I'll know where to come by listening for the music as I have a hunch the sound made with some Fostex loudspeakers will be just what I like! 😉
And for any newbies out there, take note of Graham's experience. Often pushing the probe of a multimeter onto a connection will make it OK while the probe is there giving a false impression that the connection is OK when it is not. - Status
- Not open for further replies.