• TalkBass has been independent since 1998. Add your voice.
    Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
    Join free Log in
    Want zero display ads or expanded classifieds tools? Compare plans.

GK rails in a pedal?

#182
johnk_10 said:
yeah, that sounds exactly like what the LF353's are doing (I've tried 3 of them so far). at first they seem to be fried, but if I let them sit a while they seem to work again, although they don't seem to sound as good as before they got overloaded. they do end up I think that I tried a OPA2604AP and it still did it though.
Click to expand...

I just slammed the input at C3 with a 6Vp-p signal (in spice) and none of the inputs on either opamp exceed common mode. They stay well below the +/-9V supplies, so this shouldn't be an issue.

Can you scope the inputs of U1b when this is happening? Also, can the charge pump be shutting down?

-Frank
 
#184
boomertech said:
I just slammed the input at C3 with a 6Vp-p signal (in spice) and none of the inputs on either opamp exceed common mode. They stay well below the +/-9V supplies, so this shouldn't be an issue.

Can you scope the inputs of U1b when this is happening? Also, can the charge pump be shutting down?

-Frank
Click to expand...
i'll have to do some tests tomorrow but I don't think that the LT1054 is shutting down at all or has anything to do with it it can handle 100ma. i'm taking the +9V right off the DC jack, so the -9V should have 100ma available and this circuit only draws approx. a total of 20ma at idle and up to 40-50ma when driven hard.

i'll also post a couple of sound clips of it tomorrow too.
 
  • Like
Reactions: ryansalmond
#185
johnk_10 said:
I just tried the OPA2134 and believe it or not, I actually prefer the tone of the LF353.
Click to expand...

I suggested the 2134 because it is a brute at taking punishment. Did it crap out the same way as the 353?

The data sheet for 353 again:
**********************
The amplifiers will operate with a common-mode input voltage
equal to the positive supply; however, the gain bandwidth
and slew rate may be decreased in this condition.
When the negative common-mode voltage swings to within
3V of the negative supply, an increase in input offset voltage
may occur.
**********************

We are within the 3V of the negative rail when this circuit is slammed, so the offset change at U1b could be causing the issue and fighting the U1a opamp. We can remedy this by adding another LED in series with D4. This will keep the inputs of U1b within all suggested operating parameters... even with abusive audio signals.

-Frank
 
#186
boomertech said:
I suggested the 2134 because it is a brute at taking punishment. Did it crap out the same way as the 353?
-Frank
Click to expand...

yes, but maybe a little less.


boomertech said:
We are within the 3V of the negative rail when this circuit is slammed, so the offset change at U1b could be causing the issue and fighting the U1a opamp. We can remedy this by adding another LED in series with D4. This will keep the inputs of U1b within all suggested operating parameters... even with abusive audio signals.

-Frank
Click to expand...

i think that that's a good idea. i'll try that too. I might get a away with using blue or green leds since they have a larger voltage drop than the red ones since my PCB layouts are getting bigger and bigger as it is.;)
 
#187
johnk_10 said:
i think that that's a good idea. i'll try that too. I might get a away with using blue or green leds since they have a larger voltage drop than the red ones since my PCB layouts are getting bigger and bigger as it is.;)
Click to expand...

To keep the parts count down... Instead of adding another LED, do you have 3-4V zeners? We could use zeners in place of D4 and D5 to keep the simulated HV rails below the operating conditions of U1b.

-Frank
 
Last edited:
#192
johnk_10 said:
I couldn't find my 3.6V's but did find some 5.1V ones. I tried the 5.1V's in it and right now there's no sound whatsoever, although it might have a different problem since I've been unsoldering and soldering a lot of components on it during testing.
Click to expand...

5.1V might be too much.

Hey John when i was taking my shower today I was doing some math in the steam on the shower door with the gain elements. What I am thinking now is that U1b with its open loop gain making the error corrections around the output transistors doesn't need to be so high, because the gain from U1a and the Q3/Q4 driver should be more than enough. So if you were to put a 10K from pin6 to pin7 on U1b we would take the open loop gain away from U1b and give it a gain of approx 30. I ran some simulations and the signals look real good with this gain.

-Frank
 
#193
it was another blown LF353.
this is getting to be pretty frustrating. I've now blown 6 LF353's and 1 OPA2134! i'm running out of LF353's so hopefully we can get this solved soon.
IMO, the circuit isn't stable enough yet to accept high level inputs (like having a booster pedal infront of it) without it frying the chip.
 
#194
johnk_10 said:
it was another blown LF353.
this is getting to be pretty frustrating. I've now blown 6 LF353's and 1 OPA2134! i'm running out of LF353's so hopefully we can get this solved soon.
IMO, the circuit isn't stable enough yet to accept high level inputs (like having a booster pedal infront of it) without it frying the chip.
Click to expand...

I think we should go back to the LED's and try that 10K across U1b. Before giving it any signal, please take DC voltages at the test points.

-Frank
 
#195
boomertech said:
5.1V might be too much.

Hey John when i was taking my shower today I was doing some math in the steam on the shower door with the gain elements. What I am thinking now is that U1b with its open loop gain making the error corrections around the output transistors doesn't need to be so high, because the gain from U1a and the Q3/Q4 driver should be more than enough. So if you were to put a 10K from pin6 to pin7 on U1b we would take the open loop gain away from U1b and give it a gain of approx 30. I ran some simulations and the signals look real good with this gain.

-Frank
Click to expand...


okay, I can tack a 10K in there.

BTW, I've solved a lot of mystery faults while repairing amps while stepping away from them & thinking about a solution in the shower. ;)
 
#199
johnk_10 said:
tired the 10K across U1B 6&7 and it didn't seem to make any difference.
I also blew another LF353 (for a total of 7 of them) when I tried a booster in front of it. I only have one LF353 left so I hope we can solve it before that one is fired too.
Click to expand...

Before installing the opamp let's get some DC readings. Then use 072, 082 or 062 for testing... they simulate just like the 353.

We need to see if Q1 or Q2 is shorted and feeding power into the output of the opamp.

-Frank
 

Latest posts

Top Bottom
Back