As you can see in this post here: http://www.diyaudio.com/forums/audi...-kit-building-instructions-5.html#post1524877 the offsets depend on chips and actual values are all over the map. It would be actually quite surprising if they were exactly the same for both channels.
Hi Peter,
Can you share your dimension and predrilled position/size, the U channel is a fabulous idea.
Thanks
Bi Amping & Transformers
Hi Peter
I have a dual mono GC. But now I am going to bi amp my speakers so I was planning to build another dual mono GC to bi amp my 2 drive unit speakers (one transformer per drive unit
). Do I really need to do this? - has anyone compared 2 transformers versus 4 transformers in this type of bi amp set up. Any feedback would be appreciated. I am happy to invest in the additional transformers but it seems a waste if they do not provide a significant improvement.
All the best
P
Hi Peter
I have a dual mono GC. But now I am going to bi amp my speakers so I was planning to build another dual mono GC to bi amp my 2 drive unit speakers (one transformer per drive unit
). Do I really need to do this? - has anyone compared 2 transformers versus 4 transformers in this type of bi amp set up. Any feedback would be appreciated. I am happy to invest in the additional transformers but it seems a waste if they do not provide a significant improvement.All the best
P
"Interesting" heat-problem
Hi all!
I am currently building two LM4780 kits, one bridged amp, and one parallell amp. I am currently having a rather interesting heat problem with the parallelled one.
I've built the amp with a 400VA 24v transformer, and one rectifier board. I've used star grounding as described in the building instructions thread. I've used a large bank of smoothing caps, with bypass caps and snubbers. I also have supply bypass caps on the underside of the amp PCBs. The power supply is as far as i can tell working properly. Rather perfect if i can say so myself
I initially got a heatproblem with one of the two channels. With input grounded it would produce normal DC offset values (22mV). It would get "uncomfortably hot to touch" though, when sitting on a heatsink with a rating of 0,8 C/W! The other channel would only get "slightly heated", and it also had good DC offset readings.
Thinking it was an oscillation problem, i installed a zobel network, and i also installed HF filtering caps at the input pins, as described in the datasheet of the LM4780. A 220pf capacitor between IN+ and IN-.
Thinking everything would now be perfect, i hooked it up to a set of old speakers. Both channels played beautifully, but the one channel still got HOT.
So i powered down, and started the search for anything else that could be causing the heat problems. I checked all solders, cleaned of the board with nail polish remover, measured resistor values, and even lifted one leg of the 0.1 ohm resistor, to make it stereo, in case there was a "fighting each other" issue. Still that one chip got "to hot to touch for more than 3 seconds" when sitting for a while with the input grounded!
I am now out of ideas of what might be the issue. I have a bit of hum, but i think this is a grounding issue with the signal, so i think this is a easy fix. Also the signal wires are long, and run alongside the transformer, which offcourse doesnt make it better.
As far as i know, i can now rule out: Supply issues, high frequency oscillation, ground loops, "chip to heatsink" V- problem , improper resistor values, unbalance between outputs when in "stereo-mode" (both separate outputs had good DC offsets), and dirty and bad soldering.
If any of you guys out there could give me a hint on what could be the problem it would be greatly apreciated! Since one channel is working as anticipated, i am rather baffled by the other one not wanting to. I'm almost at the point of giving up.
PS: the bridged amp is also acting like a moth******r. lots of hum, and heats to "SPike-mode" in less than ten secs. I haven't yet checked into this though, but im thinking grounding issues, oscillation, or signal ground to power ground issues. This is not the main priority though, at the time.
Hi all!
I am currently building two LM4780 kits, one bridged amp, and one parallell amp. I am currently having a rather interesting heat problem with the parallelled one.
I've built the amp with a 400VA 24v transformer, and one rectifier board. I've used star grounding as described in the building instructions thread. I've used a large bank of smoothing caps, with bypass caps and snubbers. I also have supply bypass caps on the underside of the amp PCBs. The power supply is as far as i can tell working properly. Rather perfect if i can say so myself
I initially got a heatproblem with one of the two channels. With input grounded it would produce normal DC offset values (22mV). It would get "uncomfortably hot to touch" though, when sitting on a heatsink with a rating of 0,8 C/W!
The other channel would only get "slightly heated", and it also had good DC offset readings. Thinking it was an oscillation problem, i installed a zobel network, and i also installed HF filtering caps at the input pins, as described in the datasheet of the LM4780. A 220pf capacitor between IN+ and IN-.
Thinking everything would now be perfect, i hooked it up to a set of old speakers. Both channels played beautifully, but the one channel still got HOT.
So i powered down, and started the search for anything else that could be causing the heat problems. I checked all solders, cleaned of the board with nail polish remover, measured resistor values, and even lifted one leg of the 0.1 ohm resistor, to make it stereo, in case there was a "fighting each other" issue. Still that one chip got "to hot to touch for more than 3 seconds" when sitting for a while with the input grounded!
I am now out of ideas of what might be the issue. I have a bit of hum, but i think this is a grounding issue with the signal, so i think this is a easy fix. Also the signal wires are long, and run alongside the transformer, which offcourse doesnt make it better.
As far as i know, i can now rule out: Supply issues, high frequency oscillation, ground loops, "chip to heatsink" V- problem , improper resistor values, unbalance between outputs when in "stereo-mode" (both separate outputs had good DC offsets), and dirty and bad soldering.
If any of you guys out there could give me a hint on what could be the problem it would be greatly apreciated! Since one channel is working as anticipated, i am rather baffled by the other one not wanting to. I'm almost at the point of giving up.
PS: the bridged amp is also acting like a moth******r. lots of hum, and heats to "SPike-mode" in less than ten secs. I haven't yet checked into this though, but im thinking grounding issues, oscillation, or signal ground to power ground issues. This is not the main priority though, at the time.
Possible solution
I have read a couple of threads on different forums, and seems like putting a snubber network, or a cap in parallell with the feedback resistor will reduce HF oscillation if any is going through the feedback circuit. Obviously, HF oscillation going through this circuit will be "looped" and therefore consume rather extreme amounts of current.
Any idea or suggestion about what size a cap in parallell with the feedback resistor would have to be? seems like people use a 47pf one, per feedback resistor. Any idea of what the -3db rollof frequency will be with this setup? I'm thinking this will cause some noticable "interference" in the top frequencies, allthough im not sure
Any suggestions or ideas would be greatly appreciated!
I have read a couple of threads on different forums, and seems like putting a snubber network, or a cap in parallell with the feedback resistor will reduce HF oscillation if any is going through the feedback circuit. Obviously, HF oscillation going through this circuit will be "looped" and therefore consume rather extreme amounts of current.
Any idea or suggestion about what size a cap in parallell with the feedback resistor would have to be? seems like people use a 47pf one, per feedback resistor. Any idea of what the -3db rollof frequency will be with this setup? I'm thinking this will cause some noticable "interference" in the top frequencies, allthough im not sure
Any suggestions or ideas would be greatly appreciated!
Firecrackers
Hi Peter,
I just completed building and firing up (literally) your gainclone kit. I was wondering if you know what the likely damage is to an amp (one channel) that has been connected to the PSU in reverse (ie. + to -, - to +) for enough time (< 30 seconds) for the capacitors to blow and a cloud of smoke to form? Would you expect the IC (or other parts) to be toast, or is it worth replacing the caps, re-wiring it correctly and retrying it as is? If you are not laughing too hard and possibly have had this experience perhaps you could describe the most likely result.
Thanks
Hi Peter,
I just completed building and firing up (literally) your gainclone kit. I was wondering if you know what the likely damage is to an amp (one channel) that has been connected to the PSU in reverse (ie. + to -, - to +) for enough time (< 30 seconds) for the capacitors to blow and a cloud of smoke to form? Would you expect the IC (or other parts) to be toast, or is it worth replacing the caps, re-wiring it correctly and retrying it as is? If you are not laughing too hard and possibly have had this experience perhaps you could describe the most likely result.
Thanks
Just for your information I retried the amp with new caps and proper wiring; the chip blew (likely it was already blown, as you suggested). I was wondering if you would sell a single channel of your LM3875 Premium kit, minus the power supply board and parts. Would matching to my other channel be an issue worth worrying about? The DC offset on the other channel is -47mV.
I was unsure about the transformer connection to the mains. Your Plitron transformer has a dot on it's diagram to indicate the start of the winding. My Antec transformer has not - just two windings with red and black wires for each. I connected the line to red and the neutral to black; does it matter either way?
Thanks for any help you can provide.
I was unsure about the transformer connection to the mains. Your Plitron transformer has a dot on it's diagram to indicate the start of the winding. My Antec transformer has not - just two windings with red and black wires for each. I connected the line to red and the neutral to black; does it matter either way?
Thanks for any help you can provide.
Hi Peter, I built an LM3875 kit a few months and have been listening to it daily. I was using a set of old Cerwin Vega U6 speakers that are actually satellites for a subwoofer system. I decided to try out the sub with the system the other night and it made a really nasty humming sound. I triple checked the connections as the crossover goes through the sub and thought perhaps I had erred. The next day I tried a Denon receiver and the speakers worked perfectly. I then changed the connections back to the chip amp and the nasty hum again. I figured ok perhaps the positive and negative were screwed up when I built the chip amps so I tried carefully to reverse one side and the noise was indescribable. Although the two chip amps are in the same case I have completely separate transformers and wiring except for the mains. I did some research and found that the sub is what is called a dual voice coil. Is it possibly because the two mono amps aren't matched close enough that the sub does not work? It's not a huge concern as I can still use my other amp but I would like to find a solution as I plan on building another dual mono for my main system which also uses a sub/satellite configuration. I hope I have given enough information to get some replies. ( : Thanks in advance.
