McMeat debug: Envelope ain't working

[Coriolis]

Yes, it's another one of those threads.

Been in the forum archives for 2 days now, been subbing opamps, reflowing solderjoints, checking my wiring, component values, but no luck so far: I can't get the led to light up.

What I got:

   -Signal out of pin 14 of TL074, and with "blend" pot set to no filter, I get a clean sound from the output.
"Blend" set to full filter gives me no sound.

   -Filtered sound (static) from pins 1,7,8 on TL074, or HP, BP, LP outputs respectively (so maybe I wired "blend" wrong).
With "colour" (resonance, right?) on max, I get pretty much the same sound out of all three outputs, sounds like somewhere between BP and HP, but nothing like LP. Might hear a difference if I could sweep the frequency?

    -Got signal on pin 2 of the LM1458. Got very loud and distorted, even blatty sound on pin 1! After D4, it drops a bit in volume, but is now almost gated to nothing, like a very misbiased FF or something. Chip gets about 7.8 volts from a battery on pin 8.


What I don't got:

    -No light in the LED, no matter what I do with sw-1 or "intensity"

    -No filter sweep at all, obviously.

What I tried:

    -Applied +v directly at point "M" after R9. This let me light up the led with the "intensity" pot, and gave some sweep in the filtered sound, except from LP, where it only resulted in a slight and sudden jump in volume beyond a certain degree of pot rotation.
I abstained from diming the pot with this arrangement, since I wasn't sure if the vactrols could be damaged.

    -Tried subbing a 4558 for the 1458, but then I had no sound from pin 1 at all, and nothing after that either.


My thoughts on this:

    -The LM1458 is misbiased somehow - is it getting too little juice from the battery? Why isn't it run at half supply voltage instead of full? I thought you had to do that...?
    -Maybe I manhandled the 1458 when I put it in the socket, and messed it up with a static charge, so now it craps out... But my experience with opamps like the 4558 and TL072 tells me I can handle them with my fingers and they will still work. Thought only cmos was that touchy?
    -The fact that the sound gets blattier and more gated after D4 should tell me something of the symmetry of the clipping I'm hearing, but does it tell me what might be going wrong in the first stage of the 1458?
    -Did I perhaps fry D4?


Mark Hammer suggested, in another thread on the same subject, to measure ac in the 50-150mV range in the env circuit, but unfortunately my cheap meter does not let me do that.
Right now my only options seem to be to keep going over the board, wiring, schem and layout, till something turns up. Problem is, I've been doing that A LOT now...

Oh, I'm using this schem and layout:

http://home.planet.nl/~remke063/mcmeat.html

Help a poor filter-junkie get his fix? 

C

[chunks717]

funny, (not really)- I woke up and decided to make the McMeat sweep this morning.
I built this awhile ago and just got sick of looking at it..........................
I belive all I can do at this point is build the env. circuit seperately, and make it light some leds,
then shove it back into orig. circuit..........the filter works fine.  ALMOST sweet enough to box as is...but....

just in case, I am using the NTE "equiv" for the 1458.(nte778A)..If this is a problem, I would love to know.

[Coriolis]

Well, I don't know about NTE subs, so can't help you there...

Are you getting distinct HP, BP and LP sounds from the respective pins on the TL074, or does it all sound a bit like BP to you?
I'm considering putting some LDR's in the ldr-side of my opto's just to see if I can get some sweep that way...

Swt, Markus, Joep? Are you out there somewhere? 

C

[hank reynolds 3rd]

Coriolis ...
I had a similar problem with my clone when I first built it...I realised the attack and decay pots were wired the wrong way around...I'd try another 1458 or maybe a NE5532 first though (as I remember,I tried a TL082 and the env section didn't work too well)

[markusw]

What are your settings for attack and decay (and sensitivity)??

If decay is set too low and attack too high you won't get a sweep. But I suppose you checked all settings....

For down sweep it works better if you slightly increase decay compared to up sweep.

I abstained from diming the pot with this arrangement, since I wasn't sure if the vactrols could be damaged.

I think as long as the external LED is not flickering when you hit the strings there's no need to worry about the Vactrols
Also you can check them by measuring thr resistance while turning the intensitiy pot (even if the envelope follower doesn't work so far).

I'm considering putting some LDR's in the ldr-side of my opto's just to see if I can get some sweep that way...

Good idea! when you cover those test LDRs with your hand you definitely should hear some sweep.


Markus

[markusw]

I realised the attack and decay pots were wired the wrong way around.

basically the same category as

If decay is set too low and attack too high you won't get a sweep. But I suppose you checked all settings....

[Coriolis]

...I realised the attack and decay pots were wired the wrong way around...

Yeah, I read about your problem in an old thread, but didn't immediately think it could be part of my problem...
I mean: Could it cause the 1458 to distort it's a$$ off? Probably I need to try another chip, I guess. Fresh out of 1458's though - so will try a 5532. I wonder though: Is there a quick explanation why a 4558 wouldn't work?

C

[Coriolis]

Good to see you here, markusw!
What you said about the attack and decay: That's what I initially figured, if I soldered them in reverse, I would just have to use them in reverse and they would still work. Am I right?

Also you can check them by measuring thr resistance while turning the intensitiy pot (even if the envelope follower doesn't work so far).

Hmm...but if there is not enough voltage coming out of the env circuit? Gotta light the leds in the vactrols with something, right?

C

[markusw]

That's what I initially figured, if I soldered them in reverse, I would just have to use them in reverse and they would still work. Am I right?

Yup, would be my guess too..

Hmm...but if there is not enough voltage coming out of the env circuit? Gotta light the leds in the vactrols with something, right?

OK, my mistake...obviously skipped this line

-No light in the LED, no matter what I do with sw-1 or "intensity"

I like the idea of replacing the Vactrols with plain LRDs to see if the filter is OK.

My guess is a bad solder joint, a short or maybe a wrong resistor value around the envelope follower.

Maybe check the 1458 with a simple amplifier stage on breadborad...just to be sure..

Is there a quick explanation why a 4558 wouldn't work?

I would appreciate to have an explanation for his too Maybe it doesn't deliver enough current ...??




Markus

[Coriolis]

Yeah maybe I should check the 1458 in another circuit. I did go over resistor values and such...

One thing comes to mind though:
I etched a board from the layout in the pdf, and only then did I read in another thread that there was an error in the layout.
On the schem there is an 1M8 resistor paralleled to D3 between pin 1 & 2 on the 1458, but on the layout I was using to etch, it's straddling D4 instead. So I just took out the resistor I has already put in across D4, and tacksoldered it to D3. The joints look fine and all, but who knows.
I really can't see it being a problem though.

Wonder if there were other errors on that layout though?

C

[markusw]

On the schem there is an 1M8 resistor paralleled to D3 between pin 1 & 2 on the 1458, but on the layout I was using to etch, it's straddling D4 instead. So I just took out the resistor I has already put in across D4, and tacksoldered it to D3. The joints look fine and all, but who knows.
I really can't see it being a problem though.

I remember this issue. IIRC, it shouldn't make a huge difference (think it just somehow changed the rectifier whithout really changing it's functionality).

Yeah maybe I should check the 1458 in another circuit.

At least you can be sure that it's not the chip.

There are not that many parts left in the follower..

Markus

[Coriolis]

Ha! Yeah you're right markusw, that's the beauty of debugging isn't it! 
Will keep "chipping away" 

Thanks for your time!

C

[Coriolis]

Allright, tried subbing a 5532 for the 1458 - nothing...
Actually, with "sens" on full, if I play lightly, I get very little or no output on pin 1 - but if I whack the strings hard, I hit some sort of threshold and get that loud, blatty distortion. But still no blinking led..

And since I tried putting +v directly on the led (actually before the "intens" pot so as to limit the current to the led), I know the led works.

So: Something wrong with the rectification, or the opamp biasing scheme, or what? I don't know. Gonna keep at it.

C

[ulysses]

are you using a flat battery or a 9v+ transformer?

when i use a flat battery on mine it sounds distorted. even with a relativly full battery it still triggers a lot better with a 9v+ transformer.

are you using the project guide from pisotones? i can confirm that they have all their pots back to front in their diagrams. and that one of their diagrams is wrong. in the diagram they show connecting sweep switch to intensity pot, they have the switch connecting to pot pin 1 when it should infact be pot pin 3. their incorrect wiring bypasses your intensity pot (no resistance). so if you have your sweep (up/down) switch set to up then it will be very hard to trigger the filter and you will see no light.

in the pisotones diagram they also dont state that the sweep (up/down) switch is actually 2 positions 3 pole. so make sure you set your switch correctly. this will directly affect whether the light comes on when you strum your guitar or whether the light dims when you strum your guitar.

another thought.. you do have the led round the right way dont you??

cheers
ulysses

[lowstar]

hello,
let me chime in on this too.
my "shots in the dark" are:
x) is the position switch in full or 1/2 ? because one of the positions is for static filter so you can use it as awah with a pedal, and i thought you wrote that somewhere in the circuit you get "static filter sounds" (don´t shoot me if this was much too obvious and you had checked it already)
x) when you change the up/down switch, does the control led light up in one of the positions (i think down) ? in this position, it should dim with the signal, then go back up, and the brightness should increase with the intensity-pot.
x) as ulysses said, the pisotones site has one pot-wiring wrong, and the thing with the rotary switches is a bit confusing too. i had mine wrong at the start. it is important here that you go by the schemo for those connections, not just the diagram.

so much for now, as markus is already on the task, he understands better what happens.
cheers,
lowstar

[Coriolis]

Hi again folks - it's noon here (Denmark), so I'm back in the saddle with this one.

ulysses said some things:

are you using a flat battery or a 9v+ transformer?

There might be something there, as I'm using a battery that isn't exactly fresh - although not dead either.
I'm getting below 8v from it, so if that could make a diff, I better try a regulated psu - i have one that outputs about 10.5 v.

are you using the project guide from pisotones? i can confirm that they have all their pots back to front in their diagrams. and that one of their diagrams is wrong. in the diagram they show connecting sweep switch to intensity pot, they have the switch connecting to pot pin 1 when it should infact be pot pin 3. their incorrect wiring bypasses your intensity pot (no resistance). so if you have your sweep (up/down) switch set to up then it will be very hard to trigger the filter and you will see no light.

Yep, I'm using that guide. Guess I'll check that too (again)

in the pisotones diagram they also dont state that the sweep (up/down) switch is actually 2 positions 3 pole

I believe that's taken care of using jumpers on the switch - I tested circuit around the switch for continuity, and it seems to work as intended.

another thought.. you do have the led round the right way dont you??

Hehe...I THINK so - at least it lit up with +v injected before the intensity pot, so I assume it's correct, but you know what it's like when you've debugged something for a long time; you start having doubts about your own judgment (did I actually check this or that, or do I just REMEMBER checking it).

lowstar said:

x) is the position switch in full or 1/2 ? because one of the positions is for static filter so you can use it as awah with a pedal, and i thought you wrote that somewhere in the circuit you get "static filter sounds" (don´t shoot me if this was much too obvious and you had checked it already)

When the switch is off (no input to the env follower) I should be able to sweep the filter manually using the Intensity pot, right?
Doesn't work. I get no light in the led using any of the 3 positions.

x) when you change the up/down switch, does the control led light up in one of the positions (i think down) ? in this position, it should dim with the signal, then go back up, and the brightness should increase with the intensity-pot.

This switch does nothing either, I'm afraid...

it is important here that you go by the schemo for those connections, not just the diagram.

Yep. I've pretty much discarded the diagrams now, as I only seem to get more confused...

Thanks for the input guys, I think maybe I ought to go by the"what to if it doesn't work" and post a looong one with all the voltages, observations, and so on.

C

[ulysses]

have you jumped the send return or have you wired in send return jacks?

do you get any sound at all when the trigger is set to off?

cheers
ulysses

[Coriolis]

Back again:

Hooked it up, and got filtered sound (like a half-%^&*ed wah) with blend on full filter, and unfiltered sound at the other extreme.
I then found, to my surprise, that I could sweep the filter with the intensity pot! That was new! Funny thing tho; the led didn't light up.
The mode switch seemed to work too.

Next, I tried using my regulated psu for power. Turn it on and get no sound, but a loud hum. Check all over for loose wires, alligator clips, wrong connections; measure the voltage going to the board: 10.5v. Change back to battery power; no hum, but still no sound. Figure I've fried the TL074, so replace it. Still nothing
Suddenly I see that the 074 needs vref on 3 pins, which I hadn't noticed before - I measure and get...full battery voltage! Now, the batt I'm using puts out around 7.8v when driving the mcmeat and a small testoscillator I have hooked up, so maybe before, when I was giving it 10.5 volts, I fried it? But then I swapped out the chips for a new one...

So: On the schem, vref comes from a voltage divider consisting of r14 and r15, both 22k. Since both are the same size, I should get half of Vin, right? Measuring at the junction of r14 and 15, I get 7.8v though, so something is wrong there.

A theory:
7.8v is too low, and screws up the bias point of the 1458, enough that it's output craps out, and there is not enough juice to drive a led, but just enough to give a bit of sweep.
7.8v as vref for the 074 is too high though, since the opamp then has no room to swing between the rails, right? But is 10.5 as vref high enough to fry the chip? Is there something I'm missing here?

C


Yes I have send/return jumpered.

[markusw]

From my experience the McMeat works much better at 12V. So 7.8V really might be too low.

Maybe stick a 50-100 ohm resistor and a 100µ (or larger) cap to ground between psu out and V+ of the McMeat. Could eliminate the hum.

But is 10.5 as vref high enough to fry the chip?

Don't think so.

Measuring at the junction of r14 and 15, I get 7.8v though, so something is wrong there.

Now thats pretty wierd. So you get V+ voltage at Vref?? Are you sure both resistors have the same value?? Maybe check for a short between Vref and V+...

7.8v is too low, and screws up the bias point of the 1458, enough that it's output craps out, and there is not enough juice to drive a led, but just enough to give a bit of sweep.

Seems like a good theory

7.8v as vref for the 074 is too high though, since the opamp then has no room to swing between the rails, right?

Definitely...

Sorry, I don't have any more ideas at the moment.

Markus

[Coriolis]

I don't get it, I really don't: I fiddled around with that voltage divider that wasn't dividing, untill I realized it wasn't grounded!
An alligator clip had fallen off...

So now McMeat works! Sort of...
-The envelope flutters like hell - I'm getting loads of false triggers almost no matter what I do. I suspect the 1458 is still crapping out and outputting a lot of sputters that might cause the ripple, instead of a smooth decay.
Can't imagine the Lovetone units sounding like mine!  Any ideas about that issue? Has anyone else had problems like that?

I think I may have wired the sweep switch wrong, cause i can only get upward wah. Other position does nothing. Will have to look at that.

Hmmm...other than that, it seems to be working! Hurrah! 

Sad thing is, I'm not sure what I did that fixed it...That divider was grounded initially through the alligator clip, but it didn't work at the time.
Of course, since I've reflowed the solder on the 2 divider r's, that might actually have done the trick. Who knows?

Debugging is mindboggling....

Thanks to everyone again, with your help, I'm learning a lot from this one....

C