Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Problems zeroing e12 circuit on an M3
New Posts  All Forums:Forum Nav:

Problems zeroing e12 circuit on an M3

post #1 of 10
1/24/10 at 1:06pm
Thread Starter 
So I have been trying to get my e12 circuit to operate a little more reliably with out much success.
Symptons - I can not zero the offset from amp ground channel to pin6 of U1 and at the same time expect the circuit to close the relay at turn on. I.E. if I set the offset to ZERO when I power cycle e12 it will not turn on the relay. The closet I can get to zero is -11mV. I see about a 30mV jump when the relay turns on.

I suspect this increased delta is why on certain music passages e12 falsely cuts out due to sitting closer to the theshold than designed.

I have tried 3 different parts for both U1 and U2 thinking that maybe I have a tolerance issue in one of the opamps but no luck.

My M3 measured DC offset when idle between OG and OL is 0.6mV, between OG and IG is 0.3mV

Thanks for any assistance.
Reply
post #2 of 10
1/24/10 at 4:51pm
A few things you could try:

1. Swap R5, R6, R7 and R8 for ~30kΩ resistors.
2. Decrease the value of R9 (to around 50kΩ, for example) and increase the value of C5 - aim for the lowest corner frequency you can get (10µF bipolars work well for this application).

First try (1). Then try (2).

Hope that helps.
Reply
post #3 of 10
1/25/10 at 12:48am
Thread Starter 
Well I have nothing remotely close to 30k for those resistors. I guess I can check Radio Shack for them.
Reply
post #4 of 10
1/25/10 at 12:47pm
Thread Starter 
MoodySteve - ? regarding what step 1 will do to the circuit. I understand what that with step 2 I would reduce the corner freq which would reduce the chance that some of my "false" positives. However wouldn't both of these solutions be addressing my symptoms and not the problem? Not saying that is a bad thing but it seems that the fact that I can not zero my VG between the e12 and my M3 is problem.

I wonder if my problem is more the results of just bad luck of the draw on the Q3 and Q4 specs. Would it make sense to try and match these better?
Reply
post #5 of 10
1/25/10 at 1:05pm
Pin 3 of the comparator opamp (U2) is what is used as "ground" for the offset comparison. As you note, when the relay is on the VG reference created on the e12 moves 30mV. What I have done with mine is cut pin 3 away from the board VG and connected it directly to the amp ground. Mine have both been used on Dynalos which do not use VG on the amp, but this should still work. If you have the opamp socketed, you could try this by bending pin 2 up and connecting it via a jumper to your amp OG (or IG?).
Reply
post #6 of 10
1/25/10 at 3:46pm
Thread Starter 
Ok so after reading the e12 thread over at HeadWise I think I have found some answers to my question for moodySteve.

R5-R8 increase their values will decrease the sensitivity of the e12 circuit which could solve my issue of it cutting out during music. R9 also effects sensitivity but also requires increasing C5 to keep the corner freq low. Does this sound correct?

It also seems that having to trim off-zero by 10-15mV on the e12 to account for the change in VG when then relay clicks on is not unusual. Is this correct?

If then my adjustment of 11mV is ok but I am getting cutouts during certain music passages then reducing the sensitivity of the circuit should be enough to solve my issue.
Reply
post #7 of 10
1/25/10 at 4:00pm
m1abrams, that sounds about right. I've used the ε12 in a few headphone amps and often had to reduce the sensitivity to stop the problems of either oscillation or false triggers during music playback.

You can try increasing R5-R8 to something like 20kΩ if you don't have any 30kΩ on hand.
Reply
post #8 of 10
1/25/10 at 4:07pm
Thread Starter 
Quote:
Originally Posted by MoodySteve View Post
m1abrams, that sounds about right. I've used the ε12 in a few headphone amps and often had to reduce the sensitivity to stop the problems of either oscillation or false triggers during music playback.

You can try increasing R5-R8 to something like 20kΩ if you don't have any 30kΩ on hand.
Unfortunately even though I have built a mini3, 2 gamma1 and 1 gamma2, a sigma11, M^3 and an e24 and e12 and I always order extras of parts, I seem not have have any resistors in the range between 10kΩ and 100kΩ

Guess since none of these resistors are in the signal path it would not hurt to get a few at RadioShack cause I really do not want to pay the shipping on a handful of resistors from Mouser or Digikey. Only concern is that I probably would only be able to get 5% tolerance from RS, guess I could just hand pick a few that measure close to each other for those 4 parts.
Reply
post #9 of 10
1/26/10 at 1:56am
5% is fine for this app.
Reply
post #10 of 10
1/30/10 at 1:11pm
Thread Starter 
Dropped in some 27kΩ in R5-R8 and that seems to solve my issues, thanks guys.
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Problems zeroing e12 circuit on an M3