cmoy with 1uF input cap, DC offset from source still getting through - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 4th August 2009, 01:44 AM   #1
morfic is offline morfic  United States
diyAudio Member
 
Join Date: Jan 2009
Default cmoy with 1uF input cap, DC offset from source still getting through

I followed Tangent's guide to the cmoy amp, increased the input cap size from .1uF to 1uF, added a opamp to ground channel. Sounds great, only one thing irks me, the output DC offset was huge, so i measure the source which fed it 6mV and times the default gain of 11, i saw 66-68mV on output.

It was my understanding the Input caps are supposed to strip the DC offset from the source.
Now where this does not work, the obvious questions coming up are, how was a capacitor in line with input supposed to strip the DC offset of the source? (I don't quite see where it actually was "supposed to go")

And more importantly what a possible solution is to make sure i do not amplify the source's DC offset by removing it before amping it.

(For now i dropped to a gain of 4 to amplify the DC offset less)

Thanks,

Daniel
  Reply With Quote
Old 4th August 2009, 01:57 AM   #2
diyAudio Member
 
Join Date: Mar 2007
Location: Canandaigua, NY USA
Don't know anything about the amp, but is there a resistor to ground on the far side of the cap? That's needed to hold the input at ground by allowing bias currents to flow to ground. You can also generate strange offsets if the signal is asymmetric, but that's not usually an issue except for people who test with square waves.
__________________
I may be barking up the wrong tree, but at least I'm barking!
  Reply With Quote
Old 4th August 2009, 02:08 AM   #3
morfic is offline morfic  United States
diyAudio Member
 
Join Date: Jan 2009
http://www.tangentsoft.net/audio/cmo...angent-sch.pdf

Is basically what i have now, and R2 is indeed physically between the opamp's non-inverting input pin and the capacitor.

I built it on this board: http://www.audioxpress.com/bksprods/products/pcbd-4.htm

Which makes building this pretty easy.

When you say far side, it sounds to me like you would want the cap to be between the opamp input and the resistor?

Thanks,

Daniel
  Reply With Quote
Old 4th August 2009, 02:40 AM   #4
kaos is offline kaos  United States
diyAudio Member
 
Join Date: Dec 2005
Location: Los Angeles
If you use a bipolar input op-amp for IC1 you’re guaranteed an offset with those component values. To reduce the offset you can either replace the bipolar input op-amp with a FET input amp, or put a 220uF cap in series with R3. You’ll still have an offset, but it won’t be nearly as bad.
  Reply With Quote
Old 4th August 2009, 03:24 AM   #5
morfic is offline morfic  United States
diyAudio Member
 
Join Date: Jan 2009
This brings up some obvious questions, what component values would you say would work better.
And is there a solution that focuses more on the source's offset instead of the opamps themselves?
I do have one opamp that happens to be difet, all others are bipolar.
Since all opamps i seem to like are bipolar i would rather tackle this from a "how to make the bipolars work" angle than a "let's avoid bipolars for FETs".

My main concern is properly stripping the input DC offset. I can't really work on the SMD DAC (M-Audio Transit) So i'd need to make sure the amp does not take the 6mV from the DAC and turns it into 66mV via Gain11. (Not on 11 anymore, and may drop from 4 to 3 soon, but in theory i'd like to be able to choose gain on usage, not based on controlling source DC offset amplification)

Starting from the non-inverting input, I have: Input, Resistor R2, Cap C2.

Based on Conrad's comment about the resistor being on the "far side of the cap", i wonder if i should have Input, Cap C2, Resistor R2 (opposite from how i interpret the schematic) and if there is an actual difference and a reason to go this way that i know nothing about.

Thanks,

Daniel
  Reply With Quote
Old 4th August 2009, 04:15 AM   #6
kaos is offline kaos  United States
diyAudio Member
 
Join Date: Dec 2005
Location: Los Angeles
You seem to be under a misconception. C2 does in fact block the offset from the source. The 66 mV of offset that you are experiencing is due to the finite bias current flowing out of the op-amp's inputs, which produces an IR drop across the bias resistors, primarily R2. If you want to “balance” the circuit such that you can use a bipolar op-amp with a minimum offset, change C2 to 4.7uF, R2 to 20K, R4 to 20K, R3 to 2K, and add a 100 uF cap in series with R3 @ 2K. There will still be some intrinsic offset, as there is with all op-amps (I would expect under 10 mV), but the circuit's contribution should be minimal.
  Reply With Quote
Old 4th August 2009, 01:46 PM   #7
diyAudio Member
 
Join Date: Mar 2007
Location: Canandaigua, NY USA
Not a great choice of component values for a bipolar amp- do what kaos says so the input currents are balanced. That will certainly fix the offset problem. I'm also not sure why the junction of the two batteries isn't connected to ground. The two 4.7k resistors in the power supply are just a recipe for running the batteries down faster.
__________________
I may be barking up the wrong tree, but at least I'm barking!
  Reply With Quote
Old 4th August 2009, 02:05 PM   #8
Davey is offline Davey  United States
diyAudio Member
 
Join Date: Dec 2001
Location: Bremerton, WA.
The batteries should not have a ground connection between them. That way the outputs will still remain symmetrical if one battery drops in voltage before the other.

Yep, a standard 9 volt battery is not going to last very long with this setup. However, the 1.9 mA drawn by the 4.7k resistor divider is only about 10% of the nominal total draw.

The parts list calls for an OPA2132 or OPA2134 FET-input op-amp. The circuit will work fine if the specified parts are utilized.

Cheers,

Dave.
  Reply With Quote
Old 5th August 2009, 02:22 AM   #9
morfic is offline morfic  United States
diyAudio Member
 
Join Date: Jan 2009
I now realize, i was so focused on the input DC offset, i never mentioned, the schematics only show the amp schematics, the shown basic rail splitter is not used.
Based on this: http://tangentsoft.net/elec/vgrounds.html

The railsplitter actually looks more like this:
http://tangentsoft.net/elec/bitmaps/...nd-channel.png
Minus the buffer.
I never built the cmoy version, i had the TLE version from the beginning and added the opamp later.
After some scrubbing with alcohol and measuring the amp on the desk instead of juggling it i get 7mV and 15mV offsets. With the opamp i care about the most, others show dramatically more DC offset.
The only FET (actually difet) opamp is majorly distorting in this configuration (OPA2107 i happened to actually have 2 of).

With a DC offset below 20mV i will use the amp until i have time this weekend to change the component values this weekend. I may order what i only had enough for one amp of to leave this one alone and create a second version of it.

http://www.tangentsoft.net/audio/cmo...ts.html#op-amp suggests the two burr browns since they are well behaved as far as distortion and oscillation goes, since there was no comment about fet vs bipolar i followed the advice to get sockets if i plan on trying a few.

But going through the links i revisited this link: http://www.tangentsoft.net/audio/hs-opamp.html

which talks about a lot of the things described in your replies.
(obviously i wish i had read about cranky opamps before, but the LME49720 never showed any signs of crankyness anywhere used, so i left it "for later" since i'd like to try the LM6172 in a circuit made for it)
And looking at the Input Bias Current as culprit, i now see why my LME49720 actually works as well as it does, it only has 10nF as input bias current, instead of the currents in the uF range that you all probably expected when i said bipolar. (The LT1364 that i got free with a hybrid amp (together with the OPA2107) actually lives up to the bipolar reputation with huge

Thanks for making me revisit many more pages tonight.

Only one more thing, all the formulas are considering R2, R3, R4, do i still need to worry about the pot?

Thanks for your help,

Daniel

P.S. the DMM has been a little flakey and it felt like i get time of day reponses on mV setting too. (I Can repeat the left and right offset values a couple times in a row, then i get something strange, then it gives me the expected values again. So i was wondering if someone could suggest a reliable DMM, the Extech EX330 seems like it would fit my needs, i would like to add ability to measure capacitance (am wondering if there are affordable DMM measuring down to pF range?) and am not sure i would need the True RMS offered on other models (Only time i would care about a good AC Voltage reading is to check output voltage which i would do with a 1kHz sine, and from what i understand averaging only works on sine waves, so True RMS would give more accurate readings on other but sine wave AC Voltages, and i do not see a use for me (correct me if i miss some obvious use).)
  Reply With Quote
Old 5th August 2009, 03:04 AM   #10
diyAudio Member
 
Join Date: Mar 2007
Location: Canandaigua, NY USA
I know what they're trying to accomplish with the resistive split, but IMHO you're far better off with fresh batteries and a grounded center point. Almost nothing on a modern opamp is affected by non-symmetrical power supplies, which is the reason you rarely find tracking regulators anymore- outside the audio world of course.
__________________
I may be barking up the wrong tree, but at least I'm barking!
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
cmoy dc offset correction magnade Solid State 0 5th August 2009 04:22 PM
cmoy help (DC offset) erictooth Chip Amps 7 3rd July 2009 01:32 PM
DC Offset on the input of a Threshold SL10 hifi-hospital Solid State 10 25th February 2007 09:26 AM
Question about DC offset from source. theAnonymous1 Solid State 2 12th November 2006 08:07 PM
36 V DC offset when source connected revell Chip Amps 12 27th February 2005 08:29 PM


New To Site? Need Help?

All times are GMT. The time now is 01:38 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2