Pandoras Box Fizzle

[Morbidpale]

OK, this got started over in the mods thread but im moving it over here.

As chords and notes decay I hear a lot of fizzle. I'm using an op07cp from Texas Instruments. I have tried lm308. I'm thinking they are real because they came from Jameco, reputable place. Honestly I've tried every op amp i have that will work.

Chuck D asked if it still fizzled if I removed the h11f1. It does go away. So...I started playing with the r15 value. If I go up to a 10k it pretty well solves the problem, I don't think at the expense of any real toan.

Am I barking up the right tree? Is there a better way? Why does limiting the range of the ldr make a difference?

 

[jwin615]

Are you 100% that you got an H11F1 and not an H11L1?
Could try a PC419 like in the original.

That said , R15 should be limiting the brightness of the optocoupler so I'd say you're on the right path

 

[Morbidpale]

Yes, h11f1 for sure.

 

[Chuck D. Bones]

First of all, JAMECO has not been a reputable supplier for at least the last decade. Search JAMECO in these forums and you will see many legitimate complaints about (the lack of) quality and customer service. While it's possible that JAMECO has recently reversed their business model, I'll never know because there are better sources. You would have better luck buying parts on eBay. Seriously.

There is a simple test to determine with 90% confidence whether an LM308 is legit. I have tested dozens this way and all of the legit ones pass.

OP-07 will not sound the same as a real LM308, but neither should cause the fizzle you describe.

The H11F1 is used as a noise gate in this circuit. When the signal level is below a threshold determined by the GAIN setting, D1's Vf, R11 and R13, Q1 turns on, the H11F1's LED is lit and the FET inside the H11F1 is conducting. The gating effect should be very subtle because the low LED current causes the FET's on resistance to be fairly high and the FET's saturation drain current to be low. The FET creates some very mild crossover distortion which could sound like fizzle as the note decays if the FET's saturation drain current is too high. I have not seen this with any of the H11F1's I've tested. Increasing R15 is may fix this problem, but it's a band-aid. It is easy to test the H11F1's performance on a breadboard using a trimpot, a couple of resistors, 9V power and a DMM.

Here in the Troubleshooting forum, it is customary for you to post high-res pix of your build so we can help you find workmanship errors. The leading causes of pedal malfunction, list in descending order of probability are:

1. Workmanship
2. Mods
3. Parts
4. PCB
5. Demonic possession

 

[Chuck D. Bones]

Oh, come on Steve, it wasn't THAT funny.

 

[Morbidpale]

Right, I'll get some pics when I get back to my work bench.
I do however have the same fizzle on the pcb and the breadboard (both are in action right now). Im deep into some mods but the problem existed from the very start.

What is the test for lm308's?

How do I test the h11f1?

I've never ruled out demonic possession but priests dont answer my calls about pedals anymore.

 

[Chuck D. Bones]

LM308 test
Build this circuit on a breadboard:
LM308 pin 2 to pin 6.
pin 3 to gnd via 100K.
Pin 3 to +9V via 100K.
Pin 4 to GND.
Pin 7 thru a DMM set to measure μA to +9V.
Apply power and measure current into pin 7. Typical current draw is 300μA. I have measured as low at 250μA and as high as 360μA. Most other single opamps draw at least 1mA.

H11F1 test
Build this circuit on a breadboard:
H11F1 pin 1 to a 1K resistor.
Other side of 1K resistor thru a 5K trimpot to +9V.
H11F1 pin 2 to GND.
H11F1 pin 4 to GND.
10K resistor from H11F1 pin 6 to +9V.
Apply power and measure the voltage drop across the 10K resistor.
With the trimpot at minimum resistance, the voltage drop across the 10K resistor should be around 2.5V.*
With the trimpot at maximum resistance, the voltage drop across the 10K resistor should be around 320mV.*
There should be a smooth change in voltage drop as you turn the trimpot.
* These are rough ballpark numbers. There is a lot of part-to-part variation.

 

[Morbidpale]

Okay, big surprise...the lm308s are not real.
I am measuring 2.18 mA. That's the same number I got on op07s as well.

The h11f1 was measuring from .856v to 5v5.
Im not sure what to do with that information but there it is.

Pics included. The pcb is from arionfx. So the schematic Im working on is included. Temp power wire nutted at the jack.

 

[swelchy]

my $.02 about Jameco... I bought about 10 LM308AH from there last year and they have all been fine. I just recently completed a pandoras box with those metal can LM308AH a few weeks ago with no issues. So far the few things i've bought from there have been fine.. never know though

 

[Chuck D. Bones]

Did you run any tests to verify that the LM308Hs are legit? I seriously doubt that JAMECO did. There are no factory authorized sources for out-of-productions parts, which means there is no traceability. No paper trail from you back to the manufacturer. They are a popular item to counterfeit. Many vendors have allowed themselves to be fooled by the counterfeiters. How much did you pay for them?

 

[Morbidpale]

Yes I posted the number above. They did not meet your numbers and matched the measurement of the op07s and not too far off from an ne5534 i had within reach.
I bought them a year or two ago but i recall them being around $4-5 a piece. I bought 5.
I'll sleep just fine though. I ended building and modding a rat Im tickled with using the ne5534. It was fun getting it to sound it's best to my ear and sidestep the entire mojo part debacle.

 

[Chuck D. Bones]

Those H11F1 numbers are so far off that it can't be a real H11F1. Might as well leave it out since the effect of the H11F1 is subtle when it's working right.

$4 or $5 for a real LM308H sounds too good to be true. As a point of reference, I bought a large quantity of real LM308H last year and by the time I had amortized shipping and the PayPal currency exchange vig, my cost was $4.75 each.

I used to be a skeptic about the merits of the LM308. I found that in the Rat, Expandora and Voodoo-1, it definitely sounds different (to me, better) from TL071, OP07, NE5534 and LF351. My point is that there is a reason the LM308 makes a specific tone and it's not Mojo Hype. In the end, what matters is if it sounds good to you.

I do not recommend putting a resistor in series with the power lead of any opamp unless you also install a local bypass cap. Opamps can oscillate at audio or radio frequencies if there is inadequate power supply bypassing.

 

[Morbidpale]

I do not recommend putting a resistor in series with the power lead of any opamp unless you also install a local bypass cap. Opamps can oscillate at audio or radio frequencies if there is inadequate power supply bypassing.

Roger on that. Thanks.

As for the h11f1...I got mine from stompbox parts. Would mouser be a better bet?

 

[Chuck D. Bones]

With Mouser, you're guaranteed a legit part. I bought some on AliExpress and much to my surprise, they work great. 10 for about 3 bucks.

Just to clarify, this is how I test 'em. Connect DMM across R2. Plus lead to +9V and minus lead to pin 4.

 

[Morbidpale]

Yes. That is how I did it. Thanks for all the input.
I'll grab some parts from mouser.