SOLVED God City TTTSNB left side of High (EQ) pot “shorts” out?

H

harmaes

Well-known member
I’ve posted an issue with the God City TTTSNB EQs PCB last week relating to the LT1054.
I’ve replaced the burnt out LT1054 with a new one and the build is working again.

The left side EQ (Mesa 5-band) always worked well without issues.
The right side 3 band semi-parametric EQ has had issues before and I narrowed it down to the high pot that is causing issues.
Just at the middle of the pot turning it to the left it seems to short out somehow? The right side of the pot from middle up works fully.
It sounds like the power is taken away from the sound when turning the pot left to a specific point.
I replaced the High pot with a new B10k and was expecting this to resolve the issues of a the possibly broken pot. However it didn’t fix the issue.

I validated all components and measured continuity of all parts around the High and High Frequency pots.
I also replaced the 3x TL072CP with new ones.

I attached the part of the Circuit around the High and High Freq.
Not sure what can cause this issue and more importantly how to debug how to trace the issue to be able to fix it?
So looking for some tips or leads!

Thanks for your help!
 

Attachments

  • IMG_1576.jpeg
    IMG_1576.jpeg
    195 KB · Views: 12
Sort by date Sort by votes
I’m not sure that would really help. What’s could be causing this behavior is the most importantly question for me.
Voltages are normal on the TL1054 and TL072s

EDIT: pics posted.
 
Last edited:
Upvote 0 Downvote
check the pot for damage to the wiper. can be caused by, among other things, overtightening the mounting nuts.
If you have an analog dmm, it's easy and obvious. just connect to legs 12, or 23 and sweep the pot. should see a smooth sweep. if the needle bounces, bad pot. with a digital dmm, it's a little more difficult to see as most average over time. need to sweep it very slowly.

I'm assuming this still occurs with the pots bent up like this?
 
Upvote 0 Downvote
Check with the power disconnected if the High Pot legs are going to Ground in the problem position???
I see stray bits of Solder/debri when it's blown up larger in your pics! on the component side around the high pot area!
 
Upvote 0 Downvote
Ok, I've been experimenting with this and the issue has nothing to do with the B10k High pot. I did resolve it and I'll explain what the issue was.
I replaced the High B10k pot twice and it kept crapping out the circuit. I tried different opamps and with an RC4558P the high pot worked full range but with noise like dialing in a old fashioned radio (FM/AM) in different positions of the pot.
An OPA2134 didn't work but when I switched on the 2nd channel both channels worked.

Normally this pedal has a 3pdt on/off for all 3 parts (preamp, EQ1, EQ2) or an on/off for the preamp and both EQs.
I wanted to be able to select the Mesa 5-band preamp on it's own so I build the EQs in a separate BB enclosure with 2x 3PDT to switch EQ1 / EQ2 separate or together.
When checking the schematics of the EQ and preamp I noticed a 100R going into a 100k (to ground) to output as a voltage divider at the end of the complete circuit.
So I asked God City (and the designer) via FB if it would be an issue if I would switch in EQ1 alone direct to output without the voltage divider at the end. Shouldn't be a problem they said. Well IMO it is a problem because now that I've added the 100R/100k at the end of output 1 (EQ1) between the 3pdt there's no issue with the High pot anymore.
 
Upvote 0 Downvote
Normally both EQ circuits are directly connected to each other and active.
EQ1 is a 3-band semi parametric EQ, sort of a clone of the Aphex EQ used by Metallica among others.
I wired the I1 and O1 to a separate 3PDT connected to the input jack. This 3PDT is wired to the second 3PDT which connects I2 and O2 of EQ2 which is a Mesa 5-band EQ clone and this 3pdt is wired to the output. Similar like the Paragon which has 2 channels.

Somehow adding the 100R/100k voltage divider offboard between the O1 and the 3pdt resolved the issues I had with the strange High pot behavior.
 
Upvote 0 Downvote
There outta be a series resistor between IC1-2 and C29/C30. Not sure why GCI used an 8.2nF & 10nF cap in parallel. 18nF is a common value. Connecting the output of an opamp to the - input of an opamp with only a cap is a recipe for disaster. There is a good chance that IC8 is oscillating at a few MHz. Got a scope?

If the circuit is working properly, adding that 100Ω / 100K divider will have no audible effect. I see quite a few redundant parts in there. Maybe they were added in an effort to band-aid the oscillation problem.
 
Upvote 0 Downvote
[EDIT]

I reviewed the Aphex EQF-2 service manual. It looks like the TTTSNB is a faithful copy of the EQF-2 circuit, warts and all. There is a typo in the hand-drawn service manual schematic. That typo is corrected in the TTTSNB, however the HI EQ is still prone to RF oscillations. I'll look into this further.
 
Last edited:
Upvote 0 Downvote
Chuck D. Bones said:
There outta be a series resistor between IC1-2 and C29/C30. Not sure why GCI used an 8.2nF & 10nF cap in parallel. 18nF is a common value. Connecting the output of an opamp to the - input of an opamp with only a cap is a recipe for disaster. There is a good chance that IC8 is oscillating at a few MHz. Got a scope?

If the circuit is working properly, adding that 100Ω / 100K divider will have no audible effect. I see quite a few redundant parts in there. Maybe they were added in an effort to band-aid the oscillation problem.
Click to expand...
I don’t have a scope but I did notice strange oscillation with an RC4558 opamp instead of a TL072 on most positions of the High pot.
So should there be a resistor between 6 and 7 of IC8B?

The 100R/100k divider fixes the EQ1 issue but both EQ1 and EQ2 on is too noisy IMO. I won’t use both together and probably only the 5-band EQ2. I don’t like how EQ1 (Aphex) behaves.

I did notice that he’s using 1.1k bias resistors everywhere which I never see elsewhere?
 
Upvote 0 Downvote
Chuck D. Bones said:
I just reviewed the Aphex EQF-2 service manual. It looks like there was a transcription error when copying the EQF-2 circuit for the TTTSNB. The EQF-2 contain a couple of toggle switches in the high EQ section that make interpreting the schematic "challenging."
Click to expand...
I can imagine that it’s not a clean copy. The Mesa 5-band seems more like a copy of the Mesa (I checked the Mark IIC schematic) and it’s more useable to me.

Thanks a lot for your help anyway!
 
Last edited:
Upvote 0 Downvote
It won't be me breadboarding it. I'm not down with stacking EQs for the sake of stacking EQs. Unless you're going for a stuck wah tone, then a 10-band EQ can do pretty much anything you need. if you really want a 3-band variable freq EQ, try hooking three VarioBoosts in series. Seriously.

The EQF-2's High and Low filters have four modes based on combinations of Boost, Cut, Shelving and Peaking (bandpass). The Mid filter has Boost or Cut in Peaking mode only. The modes are selected by push button. All three filter have variable boost or cut and variable frequency. The High Filter works fine in the two Peaking modes (Boost or Cut). The Shelving Boost mode sorta works, but has oscillation problems. The Shelving Cut mode doesn't even work as far as I can tell.

The EQF-2's Low filter works fine in Shelving mode (haven't analyzed Peaking mode). The Mid filter works fine. The Hi filter in Shelving mode is an example of how not to design an active filter. Aphex got pretty clever being able to switch between Peaking and Shelving modes with the push of a button, but how often are you actually going to do that? I'm guessing that Metallica found a setting they liked for certain guitar parts in certain songs and after that it was set & forget.

One parting thought. Just because you think a circuit is working correctly, doesn't mean it is. You said the circuit was noisy. In all likelihood that noise is an artifact of the Hi filter oscillating at radio frequencies.
 
Upvote 0 Downvote
Chuck D. Bones said:
One parting thought. Just because you think a circuit is working correctly, doesn't mean it is. You said the circuit was noisy. In all likelihood that noise is an artifact of the Hi filter oscillating at radio frequencies.
Click to expand...
I asked you for your help to validate if my solution was solid because your knowledge is far above my level. :)
So I wasn't convinced that the issues were fixed. But it's good to know that there are design issues with this circuit.

For now my solution did fix the osscilating issue and with both EQs on it's working without too much noise now when fully boxed up. For instance when high freq on both EQs are raised, it will add noise and some gain as well of course. For me this is more like a set and forget pedal for both sides which gives me two different flavors of EQ. Mesa 5-band for a metal like setup and the Aphex for cleaner/funkier stuff for instance.
 
Last edited:
Upvote 0 Downvote
You must log in or register to reply here.

Similar threads

H
LT1054 overheated, broken. What can cause this?
Replies
6
Views
251
Feral Feline
Feral Feline
J
  • Question
SOLVED Low Volume Queen of Bone
Replies
5
Views
265
jplev22
J
L
  • Question
SOLVED LED BEZEL - Pro-10: Green side not working when boxed
Replies
13
Views
2K
andrewsrea
andrewsrea
M
  • Solved
Digitech HM2 issues?
Replies
5
Views
203
jwin615
J
L
  • Question
SOLVED Aion VH Drive Channel - Possible crosstalk issues
Replies
4
Views
417
lowpitch
L
Back
Top