It does, but op-amp rolling is fun. 😉
I used an OPA2132P for my first Tangent CMoy and it worked wonderfully... It also works well in this build.
I think I've more or less narrowed down the problem to adding decoupling capacitors; otherwise, I'm not sure what to do. If my DMM is accurate, both the current draw and DC offset are within acceptable levels, but I'm still experiencing higher amounts of distortion with increasing volume. Note that this has only been audible with the OPA2228 and LM4562. The OPA2132, OPA2107, and AD823 do not have audible distortion with the amp.
I'm planning to pick up a package of these 0.1 uF ceramic capacitors from Radio Shack tonight. Even though they're overpriced and probably not the greatest, it should at least tell me if the addition of decoupling capacitors will solve the problem. The layout is a few posts ^^^... Which spot should be used as ground when connecting these ceramic capacitors?
I also plan to remove the resistors in the input capacitor positions (C2_L and C2_R), since at least one spec sheet (OPA2228) recommends against using them. And, just for completion's sake, I'll solder a resistor lead in that little VIA position between V+ and V-.
I used an OPA2132P for my first Tangent CMoy and it worked wonderfully... It also works well in this build.
I think I've more or less narrowed down the problem to adding decoupling capacitors; otherwise, I'm not sure what to do. If my DMM is accurate, both the current draw and DC offset are within acceptable levels, but I'm still experiencing higher amounts of distortion with increasing volume. Note that this has only been audible with the OPA2228 and LM4562. The OPA2132, OPA2107, and AD823 do not have audible distortion with the amp.
I'm planning to pick up a package of these 0.1 uF ceramic capacitors from Radio Shack tonight. Even though they're overpriced and probably not the greatest, it should at least tell me if the addition of decoupling capacitors will solve the problem. The layout is a few posts ^^^... Which spot should be used as ground when connecting these ceramic capacitors?
I also plan to remove the resistors in the input capacitor positions (C2_L and C2_R), since at least one spec sheet (OPA2228) recommends against using them. And, just for completion's sake, I'll solder a resistor lead in that little VIA position between V+ and V-.
Ground is the same as input/output jacks ground, pot ground and TLE2426 output.
And yes, don't use ceramics in audio in any position, even if the parametrs of cap is good. Use cheap small film cap(poliester) in PS, and if you need input caps, then better polipropilene there, but you don't need one.
Zigis.
And yes, don't use ceramics in audio in any position, even if the parametrs of cap is good. Use cheap small film cap(poliester) in PS, and if you need input caps, then better polipropilene there, but you don't need one.
Zigis.
The OPA2228 and LM4562 are much faster than the OPA2132, OPA2107 and AD823's you have tried. Perhaps they're unstable and oscillating?
Adding bypass caps may help if that's the case. If not, then try a small cap over the feedback resistor to limit the bandwidth, as previously mentioned.
See http://tangentsoft.net/audio/hs-opamp.html
Adding bypass caps may help if that's the case. If not, then try a small cap over the feedback resistor to limit the bandwidth, as previously mentioned.
See http://tangentsoft.net/audio/hs-opamp.html
markiemrboo, I agree, it's probably oscillating. I did notice some oscillation between 0.0 - 0.3 mV when measuring the DC offset with the OPA2228.
So, using ceramics for decoupling is okay because that has to do with power and not the audio signal, correct?
Just to be 100% clear about what I'll be soldering... Say I'm using the OPA2228 for the sake of discussing pin configurations (since I don't know if all manufacturers number their pins the same way). I take a ceramic cap and solder one leg to pin 8 (V+), take the other cap and solder a leg to pin 4 (V-), then solder the other legs to a ground position. I've heard that decoupling capacitors should be as close as possible to the board, so where is a good place to connect the ground legs? I was considering the Alps RK097's ground pin, but I couldn't find a pin diagram.
So, using ceramics for decoupling is okay because that has to do with power and not the audio signal, correct?
Just to be 100% clear about what I'll be soldering... Say I'm using the OPA2228 for the sake of discussing pin configurations (since I don't know if all manufacturers number their pins the same way). I take a ceramic cap and solder one leg to pin 8 (V+), take the other cap and solder a leg to pin 4 (V-), then solder the other legs to a ground position. I've heard that decoupling capacitors should be as close as possible to the board, so where is a good place to connect the ground legs? I was considering the Alps RK097's ground pin, but I couldn't find a pin diagram.
infinitesymph said:markiemrboo, I agree, it's probably oscillating. I did notice some oscillation between 0.0 - 0.3 mV when measuring the DC offset with the OPA2228.
So, using ceramics for decoupling is okay because that has to do with power and not the audio signal, correct?
Just to be 100% clear about what I'll be soldering... Say I'm using the OPA2228 for the sake of discussing pin configurations (since I don't know if all manufacturers number their pins the same way). I take a ceramic cap and solder one leg to pin 8 (V+), take the other cap and solder a leg to pin 4 (V-), then solder the other legs to a ground position. I've heard that decoupling capacitors should be as close as possible to the board, so where is a good place to connect the ground legs? I was considering the Alps RK097's ground pin, but I couldn't find a pin diagram.
Hi,
Some will argue that the power supply is a part of the signal path.
Ceramic NP0 / C0G types should be fine for the signal path in the traditional sense (such as the cap over the feedback resistor).
I've tried 100nF film types and X7R (I think... certainly wasn't NP0) type ceramics in my power amp supplies and both sound just fine to me. I'd say you should be just fine with them 😛 I would personally be more worried about getting the oscillation away (assuming that's what it is) rather than how a capacitor might sound.
I'm pretty sure most, if not all dual opamp pinouts are the same. What you describe sounds correct, a cap from each rail to ground, as close to the chips pins as you can. As for the grounding of these caps.... I think ideally you would want them going straight back to the TLE output, which is ground according to the schematic. I don't think it would be ideal connecting them to input / low level grounds.
Best of luck,
The TLE's output is the left pin (according to the orientation of the board layout), right? Just double checking.
I was also planning to take Zigis's earlier suggestion of running a ground wire from the Alps pot shaft to ground... That one position on the TLE might be a little cramped with the two extra resistor leads. Are there other positions that would be suitable for the ground wire?
I installed the Radio Shack ceramic caps and tested for DC offset. It was near zero, so I tried the cheap Koss TD-60 headphones--the ones that had exhibited harsh distortion when paired with a low-output portable CD player.
No more distortion! Both the OPA2228 and LM4562 work now. Thanks for your help, guys!
I did manage to achieve some major distortion by tapping on the ceramic caps, but then I realized it was because I was making one of the leads touch a pin. This seems to be the main problem with mounting these on the underside of the board. Is there some material I could place between the board and the capacitor leads that would isolate them?
Since my problem appears to be solved, is there any reason to use the bandwidth limiting capacitor in the feedback loop, or is that mostly a troubleshooting step?
Final question... Are there any benefits to leaving the resistors in the C2 (input capacitor) positions? I had heard that it was a good idea to use an identical resistor value as R3 instead of just using a jumper, but can't remember the reason why. Maybe a balance issue? 😛
No more distortion! Both the OPA2228 and LM4562 work now. Thanks for your help, guys!
I did manage to achieve some major distortion by tapping on the ceramic caps, but then I realized it was because I was making one of the leads touch a pin. This seems to be the main problem with mounting these on the underside of the board. Is there some material I could place between the board and the capacitor leads that would isolate them?
Since my problem appears to be solved, is there any reason to use the bandwidth limiting capacitor in the feedback loop, or is that mostly a troubleshooting step?
Final question... Are there any benefits to leaving the resistors in the C2 (input capacitor) positions? I had heard that it was a good idea to use an identical resistor value as R3 instead of just using a jumper, but can't remember the reason why. Maybe a balance issue? 😛
Sorry I missed a bit on your thread.... did you now put a cap in series with the output? If so, swap it for the largest capacitance value electrolytic you have... negative pole towards speaker...
The earphones and that capacitor forms a highpass filter... the lower the earphone impendance, the larger the capacitance needs to be, to not roll off the bass early.
As for isolation, try some heatshrink tubes.
The earphones and that capacitor forms a highpass filter... the lower the earphone impendance, the larger the capacitance needs to be, to not roll off the bass early.
As for isolation, try some heatshrink tubes.
No, I'm not using any output capacitors. I added some bypassing/decoupling capacitors (two 0.1 uF ceramic capacitors from Radio Shack, since they were both small and locally available) connecting the V+ and V- pins of the op-amp to ground, as suggested by Zigis all the way back in post #5. It seems to have stabilized the amp. 
I've seen a lot of different values recommended for bypassing (10 nF, 100 nF, 0.1 uF, 10 uF). Are there trade-offs with each value, or is 0.1 uF good enough for most purposes?
I've seen a lot of different values recommended for bypassing (10 nF, 100 nF, 0.1 uF, 10 uF). Are there trade-offs with each value, or is 0.1 uF good enough for most purposes?
I was reading through Tangent's guide for working with cranky op-amps, and he mentioned that it was possible to use a 1 uF - 10 uF tantalum capacitor from rail to rail. Would this work just as well as the two-capacitor solution (and is the 16V rating enough for a potential 18V power supply)?
Since I'd solved the problem on the first amp, I built a second one last night, so I'm considering trying something different. 🙂
Since I'd solved the problem on the first amp, I built a second one last night, so I'm considering trying something different. 🙂
I'll answer my own question. 😉
I purchased a 10 uF tantalum capacitor and connected it rail to rail, making sure to connect positive to V+ and negative to V-. It didn't decrease distortion; there was no audible change after installing it.
I used another pack of cheap Radio Shack ceramics and they worked perfectly. The heat shrink suggestion worked well too, so thank you for that, Nordic. I used my soldering iron to shrink it, which didn't look or smell so great, but looks were secondary to function in this case.
I purchased a 10 uF tantalum capacitor and connected it rail to rail, making sure to connect positive to V+ and negative to V-. It didn't decrease distortion; there was no audible change after installing it.
I used another pack of cheap Radio Shack ceramics and they worked perfectly. The heat shrink suggestion worked well too, so thank you for that, Nordic. I used my soldering iron to shrink it, which didn't look or smell so great, but looks were secondary to function in this case.
Distortion
I've been having some distortion problems with a Cmoy I just built. It actually sounds more like the amp is clipping. I've built one before with no problem and I use it all the time. This time I decided to use a different OPAMP and add output caps in the place of R5. The caps I used are 4.7 uF Tantalum Caps. I though this might have been the problem but i removed them and it acts the same. I also have dual micromini Pots to adjust the channels seperatly. The OPAMP i am using is the TI NE5532P. The Cmoy powersup and you can barely hear the music except when it hits certain notes the phones pop very loudly. Any insight?
I've been having some distortion problems with a Cmoy I just built. It actually sounds more like the amp is clipping. I've built one before with no problem and I use it all the time. This time I decided to use a different OPAMP and add output caps in the place of R5. The caps I used are 4.7 uF Tantalum Caps. I though this might have been the problem but i removed them and it acts the same. I also have dual micromini Pots to adjust the channels seperatly. The OPAMP i am using is the TI NE5532P. The Cmoy powersup and you can barely hear the music except when it hits certain notes the phones pop very loudly. Any insight?
Why on earth would you use output caps on this? And why tantalum, which are god awful anyhow but really bad for coupling caps. And what are you using the dual micromini pots for? Schematic?
The micro minis are on the input in place of the pad switch. I'm using them so i can adjust the left and right channels individually, I have them mounted inside the case so that I can set my levels then leave them set and use my audio device to adjust the actual volume.(I think it pointless to have an external knob that you have to constantly adjust along with the volume from your audio device) The output caps were to act as a filter and block the DC offset. The schematic I am using is here
http://tangentsoft.net/audio/cmoy-tutorial/misc/cmoy-tangent-sch.pdf
I just replaced the R5 value with the Caps(which I have now removed) and put the pots on the input.
http://tangentsoft.net/audio/cmoy-tutorial/misc/cmoy-tangent-sch.pdf
I just replaced the R5 value with the Caps(which I have now removed) and put the pots on the input.
Looking at the schematic it doesn"t give clear indication on which ground points go to VR and which go to actual ground (-9v)....
In the schematic only R2 goes to VR ,The ground on the Pot and R3 go to actual ground (-9v)....
If you connected all VR and ground points to the same Point it could cause problems and could be the cause of the problem you are haveing....
Cheers
In the schematic only R2 goes to VR ,The ground on the Pot and R3 go to actual ground (-9v)....
If you connected all VR and ground points to the same Point it could cause problems and could be the cause of the problem you are haveing....
Cheers
After doing a lot of searching I found some info about batches of counterfeit/fake opamps. The batch of op amps i got came off ebay from somewhere in hong kong. I went to radio shack and picked up some generic op amp and they work beautifully! Another side note, I bumped the output caps up to 220uf electrolytics. I've now ordered some burr brown 2227s from digi-key. No more ebayed opamps from now on.
- Status
- Not open for further replies.