Signal ducking and then recovering

[Dar]

I've been smashing pedals together for my other half with varying levels of success and understanding for a few years now so I thought I'd try to build to an amp.

It's not an "amp" amp, it's a vero marshall JCM800 superlead pre-amp wired directly into a standalone poweramp. The power amp is just a regular 300W power amp that I took out of it's chassis and put in this one and soldered the signal wired direct to the input jack.

I used this layout at Tagboard Effects for the JCM800, everything is as it is on there.

This all worked great and he was really happy with it, until he asked for a little treble boost. I used a tilt EQ circuit for this, this one from another guitar fx forum. Everything is as is on there, except I had to use electrolytics for the 1u caps.

A basic block diagram for the amp would be:

Both the pre-amp and EQ are powered by one of these regulators/step-down convertors: https://www.bitsbox.co.uk/index.php?main_page=product_info&cPath=140_171&products_id=3202

I'm running them both at 18v (I've checked and used big enough ratings), from a 26v supply tapped from the power amp.

Now, these work and sound good together but when he clicks, on or off, a pedal plugged in the front, ones with a 3PDT switch, the signal ducks out and gradually recovers. Like a compressor with a slow release.

I've tried popping a 1M resistor across the input, just in case, and putting a buffer before the amp and neither made a change.

Any idea what could be causing the signal to duck and recover?

[iainpunk]

i'd almost say it s a DC problem, i had such a problem with an overdrive pedal once, but it had a huge input cap 33uF i replaced that with a smaller cap and that took care of the problem. (i have an active guitar with a small DC offset which was the only guitar that gave problems).

do you have a proper schematic for that tilt EQ? that might help us help you better,

cheers, Iain

[R.G.]

Iain is right. Something is suddenly changing the DC conditions when the pedal is switched in, and the recovery is the time the DC conditions take to recover. At least one pedal circuit I remember uses exactly this technique to cause an attack delay.

Iain is also right that the best way for us to help you would be an accurate schematic. The cure will almost certainly be finding the capacitor that's being dumped with the pedal switching, and fixing whatever is dragging it down.

One quick question - does it do this with all pedals or only one particular one?

[composition4]

Hey RG, can you name the pedal that intentionally uses this? Would be interested to take a look at how it's implemented.

[iainpunk]

holy s*it i forgot:

welcome to the forum

cheers, Iain

[Dar]

do you have a proper schematic for that tilt EQ? that might help us help you better,

Iain is also right that the best way for us to help you would be an accurate schematic. The cure will almost certainly be finding the capacitor that's being dumped with the pedal switching, and fixing whatever is dragging it down.

One quick question - does it do this with all pedals or only one particular one?

This is the schematic for the EQ, minus the power supply for the IC (I think):


I used 4.7nF for C3 & C4 to get a centre freq. of about 1kHz.

It does it with all the pedals with a 3PDT switch, but not the boss soft switches. Using a DIY box-of-rock or a DIY EQD speaker cranker, it seems more noticeable when turning the pedal off. (both worked flawlessly with his old peavey stereo chorus amp )

The EQ, pre-amp and power-amp are all wired together in series inside an old laney combo (which helpfully provided most of the pots for this build).  It worked amazing before I put the EQ in, but the boss demands what the boss gets! 

holy s*it i forgot:

welcome to the forum

cheers, Iain

Thanks! Long time listener, first time caller type thing!

[iainpunk]

try replacing the input cap C1 with a 47nF
that would let enough bass through to keep it the same, but it will be faster in regards to the DC conditions.

cheers,

[Dar]

Swapped it out for a 47nF and it's kinda sorted the problem!

Now the signal is steady except when turning the Super-hard-on side of the box of rock OFF? It's definitely better than it was, can I go lower with the input cap? Could the 100k resistor before it have anything to do with it? I notice it's after the cap in the schematic I posted but before it on the vero layout

[iainpunk]

i think its more use full to search where the DC comes from in the first place, a good practice to begin with.

cheers, Iain

[antonis]

It does it with all the pedals with a 3PDT switch, but not the boss soft switches.

Are 3PDT switches wired with IN AND OUT grounded when bypassed..??
(despite pull-down resistors existence..)

P.S.
Also, Welcome..

[Dar]

Thank you all for the suggestions, it's really appreciated

It does it with all the pedals with a 3PDT switch, but not the boss soft switches.

Are 3PDT switches wired with IN AND OUT grounded when bypassed..??
(despite pull-down resistors existence..)

P.S.
Also, Welcome..

Yeah, I'm a bit lazy and don't always include the pull-down resistor in my pedals but everything's grounded together. On this endeavour I've collected all the grounds together on a little bit of vero.

I'm really struggling to get my head round this, everything worked fine until I added the eq board! Apart from sharing the power supply (rated at 2A output), the only contact it has with pre- and power-amp is a wire from the input jack, and a wire to the pre-amp circuit!

EDIT: I think I've fixed it, I swapped the output cap(?), C4 for a 47nF as well instead of the 1uF marked on the vero and that matches the schematic a little better but I've no idea if I've actually fixed it or just hidden the problem!

[iainpunk]

Yeah, I'm a bit lazy and don't always include the pull-down resistor in my pedals

if you are going to be lazy, don't be lazy when it comes to pull down resistors!!!! they are more important than people give them credit for.

cheers, Iain

[Elektrojänis]

Are you actually using that circuit from a single 18 volt suply without any biasing to the opamp inputs? (I mean did you just connect the minus input of the opamp staright to ground?)

If so, that would probably do some weird things. That looks like it's meant for a bipolar supply... like +9V,  ground and a separate -9v.

[iainpunk]

That looks like it's meant for a bipolar supply... like +9V,  ground and a separate -9v.

well, 2x9=18 eh?!!! and a virtual ground is really simple to implement, especially in a circuit like this.

in concept circuits like that eq they almost always use a bipolar supply, a virtual ground to replace that is often up to the engineer at hand how to handle it, because the point of such drawings is to convey a concept or technique, not a full production ready design.

cheers, Iain

[Elektrojänis]

well, 2x9=18 eh?!!! and a virtual ground is really simple to implement, especially in a circuit like this.

But was it implemented by OP? That's the question.

OP also mentions "one of these regulators/step-down convertors". One. I might be hanging on to details too much and sorry for that, but there is really no way to know what's really happening in his amp without asking all sorts of details.

To me even the strip-board layout linked looks like it was not (even though it says verified). Strip-board and vero -layouts are a bit hard to follow though... Easy to build but hard to understand and troubleshoot.

[iainpunk]

yeah, i personally h@te vero/stripboard because it's incentivises to make sub circuits scattered through each other.

i strongly believe that OP used virtual ground because:
A] it works like i should
B] i assume they have enough knowledge to implement virtual ground based on the described past experience with pedals
C] the layout seems to be verified

cheers, Iain