Harmonic Energizer Gain pot failure stumped

[Ben79]

Hi all,

I've been working on a Systech Harmonic Energizer clone for a friend and it's all good except I'm getting no change when using the gain pot.

Thanks to Mark Hammer's posted mod, I used a 250k pot instead of the stock 50k and a 3 position on off on dpdt switch with a set of back to back diodes on one side and a 68k resistor on the other, both in parallel with the pot.  This should yield 3 gain options - stock 50k (250k and 68k in parallel), 250k with no clipping diodes and 250k with clipping diodes.

I've tested the resistance at the chip pins and it's as it should be but the gain pot does nothing with the switch in any of the 3 positions.  The circuit seems to be stuck at a rather low gain.

I've checked all the usual things - wiring, grounds, solder bridges, but reached the limit of my capabilities as I know almost nothing about opamps.  The resistance at the chip pins is as it should be.  I've also been in touch with Mark who says there should be a problem with my wiring somewhere and I'm sure he's right but I can't find it....the gain control is so simple, isn't it....?

Anyway, here's some photos of my work.







And here's the schem and layout





If anyone can steer me towards diagnosing this problem I'd be super grateful.

Cheers,
Ben

[Kipper4]

I'd start and check the 1k 10uf hanging off pin 13. (gain control) use the continuaty mode on your DMM to confirm they are soldered to each other and the cap goes to ground.
Then check the wires from the gain pots lug to pins 13 & 14.
I might be way off base.

Are the EQ filter circuits effective btw?
Good luck with it.

[Ben79]

Thanks Kipper,

I checked those continuities and they're all good.  The wires to the pots and switch I think must be fine - I just checked the resistance again across pins 13 and 14 and it varies exactly as you'd expect when I turn the pot and flick the switch.

The bandwidth and centre frequency controls work as I'd expect.

[slacker]

If the resistance measured on pins 13 and 14 of the opamp varies  between 0 and 50k or 250k Ohms depending on the pot and switch settings then the problem pretty much has to be with the 1k resistor and the 10uF cap. Is the resistor really 1k?

[kaycee]

I've just built one from this layout and it works fine. The gain control works fine as it is, I'd suggest making it as standard, verifying the build and then try modding it. It's a fun box.

[Ben79]

Thanks for the help guys.

I'm still stumped.  If the 10uf cap and the 1k resistor are good and pins 13 and 14 have a variable resistance across them then it should just work considering the circuit is otherwise functioning normally....shouldn't it?  I'm wondering if there's anything else that could be at fault that would cause the gain control not to function.

I'm stumped.

Some kind of black magic curse maybe....what do you do when your pedal is possessed by a malign spirit?

[slacker]

Yeah, If they're all good then the pot should work. It might be time to start audio probing, the signal on pins 1 and 12 of the opamp should be the same, then with the gain pot turned all the way down pins 12 and 14 should be the same, turning up the gain pot should make the signal on pin 14 louder and probably distorted at high settings. The signal on SW1 should match the signal on pin 14. Try that and see what you get, it might give us some clues.

[Ben79]

Thanks, never done audio probing before so this is a chance to learn a new debugging skill.

[Ben79]

I did some probing.

1 and 12 were the same.

12 and 14 were the same but no difference occured to the signal at pin 14 when the gain pot was turned up.

Signal at sw1 matched that at pin 14 (though the horrendous noises I was making when making contact with my improvised probe made it tricky).

All as you'd expect given what it's doing, right?  I'm gonna have to pull those components around the gain pot and test them cos I can't see what else it could be now.  They're definitely all soldered fine and the right value but maybe I've got a dud in there.

Is there any chance it could be anything to do with the FET?  I've used an MPF102 instead of the 2n5457, but that wouldn't explain the gain pot having NO effect would it?

UPDATE: I found a 2n5457 and that made no difference (clutching at straws there).  I pulled the 10uf to ground from the gain section and tested it....it was reading 118uf......'Yesss! Problem solved!!......oh, battery's dying in my meter and the cap is actually fine.'

I've obviously made mistakes before but this has gone on and on and I'm completely perplexed.  The frustration is killing me but I'm trying to channel it into solving the problem. It's gonna be ok isn't it.  One day, it will work.

[anotherjim]

Sometimes, if the pins under suspicion seem ok when resistance checking, it could be a condition that is only present when power is on, and that suggests a short to another part of the circuit could be the problem.
E.g. an opamp pin (except power) can be high impedance with power off and won't show on a resistance test  - unless you check resistance between each chip pin and every other chip pin. This is quick & easy to do. Explanation takes longer...

Set the DMM to diode/beep. Start red probe to pin1, black to 2, then black to 3... until you've tested pin 1 to every pin to 14
Then red probe to pin 2 and black to 3 & run round up to 14 again. Keep moving the red probe on one pin every time the black has reached the highest pin. There is no need to probe pin numbers below the red probe's current pin with the black, they have already been tested.
You only have to stroke the black probe over the other pins as you go round while listening for a beep. When you do hear a beep, check with the layout & schematic that that is a case where those pins are connected  - you should find several on your board - 1-12, 3-5 etc. It doesn't actually matter which way around the probes are, but it helped the description to give them names.
When done you will be sure there are no shorts between the chip pins except for those intended.
I always do this before powering up a new build.

Some of your wire ends are pointing close to adjacent tracks, though it could be the camera fooling me. I like to be able to run a scalpel/xacto blade between the tracks on stripboard without the blade snagging. That proves it's reasonably clear and most likely will clear any hard to see shorts that were there.

[Ben79]

Thanks Jim,

I've had a go at what you said with the unit off and on.

Off, I get the following...

1+12
3+5
5+10

On, I get this...

1+4
1+2
1+7
1+8
1+12
1+14

2+12
2+14

3+5
3+10

5+10

7+8
7+14

8+12
8+14

11+12
11+14

12+14

*7+10 made a short beep that fizzled out over about a second as if the DMM was dying (it wasn't)....never heard that before
1

I don't know enough about opamps to know what all that means but I can see that the continuities with the unit off are consistent with the layout.

I've tested all the tracks several times for shorts with the DMM and found none.

[anotherjim]

Argh! No Ben - don't do continuity or resistance tests with power on. Your DMM can probably survive that with a low voltage circuit, but could be fried on anything else.
Those 3 connections you found look ok, so you can rule that lot out.
With power on, every op-amp pin except 4 & 11 should be at Vb voltage - is that so?

As already pointed out, the gain amp relies on connection via 1k & 10uF. If that path is open-circuit, the amp with be a unity buffer whatever the gain pot setting. Tricky to test a cap with DMM but it is possible if you use a high resistance range and see if it takes some time for the reading to reach infinity and then reverse the probes and watch it fall to zero and then back up to infinity (power off!) - but that test can be ruined by paths thru the rest of the circuit with the cap in circuit. Easier to try changing the cap  or tack solder another one on the back of it - or...
...if you can remove the cap, temporarily connect the 1k end to Vb and see if the gain pot works then.

[Ben79]

Haha! I promise (or at least hope) I'm not as stupid as that has made me look.  I guess the DMM detects continuity by the presence of current....of course there's current if the thing is on!  What a dufus 

I've swapped the cap with another 10uf (non-polar) electro which tested good before I put it in.  No difference.

I'll check the voltages in the morning.

Thanks again for your help.

[Ben79]

Got the voltages....

All pins are at 4.55 except...

4. 9.04
11. 0

These are obviously the 9v in and ground so I know they're ok.....and that would mean all others are at Vb so that seems fine too.

[whoisalhedges]

I'm afraid I won't be of much help, I'm pretty weak on op amps myself - but I have to know where you got those orange and purple electrolytics!

[Ben79]

Like everything else I get, I would have bought those on Ebay UK, probably as part of large lots from someone offloading their surplus. 

[Ben79]

I got the gain pot working!

I had tested all the joints in the gain circuit for continuity several times with the DMM but I reflowed them anyway and it suddenly started working.  I guess I learned my lesson that just because the meter says it's good it might not be good enough or perhaps the pressure from the probe was making a bad joint temporarily good again.

So that's working.

The issue it has now is some crackly high end distortion when I hit the string any more than softly.  It's not a pleasant opamp overdrive or diode clip, something unpleasant - perhaps another bad connection somewhere  although it runs fine aside from that.  The crackles are there at any gain setting and even when I back off the guitar volume.

Also gonna have to install a volume pot at the end with that modified 250k (with clipping diodes) gain control.

[Mark Hammer]

I got you into a whole heap of trouble, didn't I?   

[Ben79]

Yes you did 

Though I do now have some basic idea of how an opamp works thanks to looking into this circuit, which is great - I'd only ever worked with transistors before.

I hope getting to the source of this nasty distortion doesn't take me another 6 weeks....