Adding low pass filter to Fortin 33 copy (high freq noise/hiss problem)

[nothing_nothing]

My Fortin 33 copy adds weird very high frequency noise/hiss spike around 16-18kHz, some amps filter this hiss, some don't and it sounds very "digital", hissy, hi-fi, it's like putting graphic EQ in the amp's loop and cranking 16 kHz slider all the way up, or maybe even worse :|

I did some research and decided to add low pass filter around 10kHz to my pedal using 10 kOhm resistor + 1.5 nanofarad capacitor and seems like it works partially... It removed some of the hissing noise, but there is still huge spike around 16-18kHz...

Wihout Low pass filter:



With low pass filter:




It looks like this when I plug this pedal directly into my audio interface (no amp/cab sim):



I can hear this high frequency crap comming trough some of my amps :|

Do you have any ideas what is causing it? I have no pedal building knowledge, it looks like rocket science to me, all I can do is medicore soldering :|

[MikeA]

Welcome to the forum!  Thanks for all the research and details you provided.

That spike you're seeing around 18 kHz is from the charge pump that's bumping up your 9V to something over 30V.  It's the IC, diodes and caps grouped at bottom right on the board.  In my experience, even if the power bus is filtered, the caps radiate the charge pump's switching frequency, and here they're right beside your audio op amp.  Best practice would be to filter the power bus before the IC and/or move the assembly to a separate board a bit farther away from the audio circuit, when designing this kind of circuit. 

You calculated your filter correctly, the knee would be 10.6 kHz, but the rolloff is not very steep above that for a passive RC filter, as your ears tell you.

Some amps pass this noise thru more than others (I have a tube amp that makes this range very audible, and a solid state amp that almost completely suppresses it.) Wish I could be more helpful on this!   

[nothing_nothing]

So is there anything I can do to simply kill/reduce this noise? Like remove/add something to the circut? Simple fix without messing too much with my pedal?

BTW adding low pass filter at the input of my amp completly kills this noise when using my boss katana, but I can't do it with my tube amps :/

[Vivek]

That Fortin 33 is an interesting circuit

Being just a TC integrated preamp but with fixed tone controls

That TC integrated preamp had very interesting schematic, the best guess being that the transistor in front somehow reduced noise of noisy opamps of decades ago

One of the drawbacks of that odd schematic is the need for high voltage

Modern opamps are low noise, high impedance, rail to rail

The Fortin 33 could be redesigned to work with 9V supplies, and output a healthy 4vp output

Or use a low cost buck boost convertor that switches in high MHz frequency range.

[MikeA]

So is there anything I can do to simply kill/reduce this noise? Like remove/add something to the circut? Simple fix without messing too much with my pedal?

Here's an easy fix you can try.  Cut a strip of aluminum foil a bit wider than the height of caps C9 and C11, and about 2 inches long.  Wrap a piece of tape completely around it, both sides and all the edges, so there's no aluminum showing.  (Any non-conductive tape will work - packing tape, cellophane tape, freezer tape, etc, if in doubt check it with your ohmmeter.) Curl it into a cylinder.  With power on and listening to your noise, lower it down over C9/C11 like the circle in the pic.  (You can also try C10/C12, they're less likely to be noisy but a lot closer to your op amp.)   If you're lucky, one or both of these may reduce the switching noise enough to live with.  If it works, secure it in place with a dab of silicone caulk or hot glue.

[Vivek]

Is the BOOST pin of the charge pump connected to Vcc in your clone ?

[iainpunk]

im having a hard time understanding the functioning of the circuit. the bias of the opamp is at 10/11th of the Vcc, the transistor that directly goes into the opamp has its base also biased at the same voltage, really close to the supply rails, so i would suspect early clipping, but that doesn't seem to be the case...

its also filled with high value resistors, if i were to re-do the design, id use lower value resistors. might lower the amount of noise as well

cheers

[nothing_nothing]

Is the BOOST pin of the charge pump connected to Vcc in your clone ?

Ummm... That black square is charge pump I guess, but what is Vcc?

[Vivek]

Could you post the schematic that you used. Thanks.

[Vivek]

im having a hard time understanding the functioning of the circuit.
cheers

https://www.diystompboxes.com/smfforum/index.php?topic=122749.0

Check out Sir Rob's calculations here.

[soggybag]

What vivek said. Check the charge pump. Some have a frequency boost that moves increases the frequency they operate at above audio range. You may need to engage this. On some chips it's just connecting pins 1 and 8. Check the documentation for your charge pump.

[nothing_nothing]

This pump is called TL071, I have heard that adding 100pF capacitor to this pump between pins 2 and 6 can fix my problem, can anoyne confirm this?

[bluebunny]

The TL071 is a single op-amp, not a charge pump.  The other IC isn't clear enough for my eyes to read all of its markings.

[nothing_nothing]

Guys, I made a small progress.

If I add 1.5nF capacitor to the TL071 (between pins 2-6) the noise is 99% gone.

Also this second black square thing is called TC1044SEPA, if I connect pins 7-8 or 7-6 together, or if I connect pin 7 to the ground on this thing, noise is completely gone, but my tone turns into farts if I pick my guitar strings too hard :|

I would say that first method is a solution to my problem, but is it safe? Also different cap may work better for this job?

[Vivek]

We are all in the dark since you did not post the schematic that you used.

There could be atleast 3 ways to solve your issue.They can be used simultaneously or individually

A) Reduce the gain of the gain of the Opamp at high frequencies by having a cap in the feedback loop

B) Having a low pass filter that cuts out hiss etc

C) Increase the switching frequency of the charge pump by using the BOOST function

The spec sheet and typical circuit of the Charge Pump is here:
https://html.alldatasheet.com/html-pdf/26002/TELCOM/TC1044SCPA/184/1/TC1044SCPA.html

[nothing_nothing]

The problem is I bought this pedal from second hand, I have no schematic

[ElectricDruid]

In my view, the key point from the TC1044 datasheet is that the "Boost" to 45KHz is engaged by connecting pin 1 to +V. So check with your multimeter that that has been done. If not, that'd be an easy fix.

Honestly though, it seems a bit obvious to *not* have been done, but you never know...

HTH,
Tom

[mdcmdcmdc]

I don't think the OP's board is from PPCB, but it's probably based on the same trace/schematic:

[ElectricDruid]

Jeez, that schematic looks like a university level "work out the equivalent impedance of this network" problem!!

Who the hell puts 820R+1K8+220R in series? It really has to be 2840 ohms? 2K7 or 3K not good enough?!? Or you could have 2K87 if you really insist on going to E48!!

The first thing I'd do if I was "cloning" such a thing would be some extensive rationalisation!

[nothing_nothing]

Ok guys I did some testing...

Connecting pin 1 to +V on TC1044 does nothing.

Best solution I foud is connecting pins 2-6 on TL071 using 1.5 nF cap, I wonder if using 2.2nF, 5nF, or even 10+nF cap will reduce this noice even further?