Build characteristics of Pete Cornish

[r080]

I posted this in the other forum, but was curious to get some other perspectives. There are a couple things I notice in Pete Cornish pedal gut shots and schematics that seem to be common in every build. I am looking for input on what the reasoning behind some of them could be to better decide whether they are worth the effort to do in personal builds.

Shielded cable everywhere inside – A bunch of connections use shielded wire (pots, board to board, board to output), but with fairly long unshielded pigtails. (The input goes straight to the buffer with a short wire. That looks great.) This looks like a lot of extra effort.

Output jack not connected to the enclosure, small capacitor at from output ground to chassis – I think this is called Hybrid grounding, and makes sense for RFI rejection when your products are used by professionals that simply must never pick up the local talk show station.

Slightly more complicated power supply – this makes lots of sense, and is probably just good engineering.

Rivets for the enclosure ground connections – wouldn't nut/bolt/star washer be better? Is the aluminum rivet to prevent any dissimilar metal issues?

The two resistors on the output jack – Could this be to prevent popping when plugging into the output? I don't have anything built this way, nor do I own any Cornish pedals to test it.

[swamphorn]

From the pictures I've seen inside Cornish's pedals it looks like the off-board connections are in thick, parallel bundles--the worst case for parasitic coupling between wires. Shielding the cables mitigates this effect. So does keeping your cables short and at right angles but the devil you know is better than the devil you don't; bundled, shielded cables are more predictable than random shielded ones.

I don't know if I'd call his power supply schematic more complex; a Schottky diode is pretty standard fare for reverse polarity protection. Hybrid grounding is a new concept for me. It seems like the purpose is to shunt high-frequency noise to ground while preventing low-frequency and DC current from flowing along the cable screen. This is done by DC grounding the circuit at one point and AC grounding at every other point. Ostensibly this is very desirable when grounding balanced connections but perhaps it also helps for unbalanced connections. As for the rivets, I would imagine it mitigates any risk of a washer coming loose and rattling around in the pedal (but as you stated, a star or other locking washer would minimize this risk--but perhaps Cornish is more about eliminating it). I can't speak for the output jack resistors; I can't find any schematic that includes them.

[amptramp]

I don't have much faith in aluminum rivets making good connections to an aluminum chassis.  Even stainless star washers would only have about a quarter volt difference in the electromotive scale with aluminum.  Aluminum has a hard shell of aluminum oxide around it and the use of rivets does not guarantee that any contact will remain conductive.  Star washers would bite into the aluminum and retain conductivity longer.

[willienillie]

Star washers would bite into the aluminum and retain conductivity longer.

And can be easily tightened when needed in the future.  I don't care for rivets myself.

[Rob Strand]

Output jack not connected to the enclosure, small capacitor at from output ground to chassis – I think this is called Hybrid grounding, and makes sense for RFI rejection when your products are used by professionals that simply must never pick up the local talk show station.

I'd say an isolate jack with a cap is probably a little *worse* than a non-isolated jack.   Generally the idea behind using isolated jack is to enforce single point grounding and avoid hum loops.  So if you look at the options for the output jack:

Boss style:  Non-isolated jacks.   The output jack ground is *not* wired.  It picks up the ground connection via the chassis which comes from single point ground at the input jack.    The pro's are good RF shielding and no *internal* ground loop.  The cons are bad audio connection if the jack nuts come loose.

In the Boss case, if you were to *wire* the output jack back to the input ground to avoid the bad connection issue then there is a ground loop because of the grounding through the case and the wire.

So the next step is to replace the output jack with an isolate jack.  So at this point the external cable connected to the output jack can act like a antenna and pick-up RF (it can also conduct RF along it).  That RF will now be present on the wire you connected from the output jack ground to the input jack ground.  So that opens up the possibility of the small wire effectively re-transmitting the external RF back into the enclosure.  Completely defeating the purpose the enlosure's RF shielding!

So the way around that is to bypass the RF coming into the enclosure along the ground wire to the enclosure ground.   The way that is done is to put a cap between the output jack ground and the enclosure.   That's exactly what Pete Cornish does (the trick is done in a lot of professional equipment, both audio and test gear).

So you have to ask what problem has been solved over the Boss method?
- The dodgy audio ground problem
- A second advantage is the audio is not passing through the chassis.   That's only a minor advantage for a sealed diecast box with two connectors like a pedal.
-  If you want to split hairs the Cornish method puts RF on inside surface of the enclosure where as Boss keeps it on the outside, which is better.  So the Boss method wins provided the external RF doesn't create a potential between the input and output socket ground.
- Boss method won't be 100% immune to low frequency currents flowing down the ground of all the effects.

So the output socket ground isn't a big difference.  Both the Boss method and the Cornish methods are OK for pedals.  The in-between methods are probably not.  The Boss method is ultimately more economical.

[Fenton Bresler]

The resistors on the output jack look like a small series resistor followed by a larger one to ground. My guess is there are no such resistors after the cap at the output of the effect board or the buffered bypass board, and the ones on the jack are serving double duty. Saves two resistors and the board space for four.

EDIT: I've just looked at Aion's trace of the SS-3 on the other site, there is a resistor to ground followed by one in series at the output of each board. So just what is the purpose for these resistors on the output jack?

[bool]

...
The Boss method is ultimately more economical.
...

And that, my friends, is a huge selling point ... "everything counts in large amounts".

[Rob Strand]

I've just looked at Aion's trace of the SS-3 on the other site, there is a resistor to ground followed by one in series at the output of each board. So just what is the purpose for these resistors on the output jack?

When I look at Aion's schematic that part of the circuit gets switched out.  Like that, it didn't make sense.

However, when I look at the gut shots the Pete Cornish pedal those parts are mounted right on the output socket and it does makes sense.  By putting a series resistor right at the socket it helps prevent RF (from outside of the pedal) which is on the center lead from getting onto the output wire (that's inside the enclosure) and re-radiating inside the enclosure.  The resistor to ground probably doesn't achieve much.

Overall I think Pete Cornish's approach is to try to do as much as you possibly can to avoid all possible problems.   At home you might get away with many short cuts and bad practices and never realize you even have a problem.   What he is doing is trying to do *something* for all possible evils out in the field.    You could say he is going overboard but with his high-end clients and years of field experience I'm sure he's seen some weird problems over the years.   What you see is his recipe how solve all those problems.   He must be happy with it if he keeps that pattern across all his effects.

On the other side of the coin there's a lot of Boss pedals out there and they have kept their recipe for many years as well!

[r080]

I don't know if I'd call his power supply schematic more complex

I am referring to the low value series resistors isolating power for different sections, each with their own filter caps. I don't see very often in effects schematics. Complex was probably not the right word.

I can't find any schematic that includes them.

Dirk Hendrik's G2 schematic is a good example. There is a 100 ohm resistor across the normalled tip such that, when unplugged, the resistor is shorted. The output end of it is connected through a 50k to ground. When plugged in, the resistor is in the circuit. It appears the moment you insert an output plug, it is connected through the 50k to ground. The next moment, the jack switch opens, putting the 100 ohm in the path. Rob's statement about RF would make sense (kind of like a Ferrite bead?), it just seems strange for it to be on the switched jack, rather than just in series. Maybe that was the most convenient place to get it near the output.

[mozz]

Aluminum and the rivet could be irradited and would not corrode.

Edit, irridited.

[r080]

What sort of radiation would do that? I have heard of coatings you can use, but I don't know if we would see evidence of that around the rivet.

I have a feeling if Pete Cornish used something like radiation, we would see mention of it in the ad copy.

[amptramp]

What sort of radiation would do that? I have heard of coatings you can use, but I don't know if we would see evidence of that around the rivet.

I have a feeling if Pete Cornish used something like radiation, we would see mention of it in the ad copy.

I think what mozz meant to say was irridited, which is a conductive surface conversion coating like alodyne.  It uses chromic acid to go through the aluminum oxide and form an aluminum chromate coating.  CrH₂O₄ (chromic acid) is hexavalent, so it is somewhat poisonous and if you remember the movie "Erin Brockovich", she was instrumental in getting Pacific Gas & Electric to pay hundreds of millions to people who were sickened by hexavalent chromium that got into the water table.  It is conductive and is an excellent paint primer for aluminum, two uses that keep it going.

[r080]

That makes a lot more sense than the way I was reading it.

[Rob Strand]

The aion pics show a star washer between the earth ring and the aluminium.  It also looks like something has been done to the aluminium in that area, perhaps it was ground clean.

hexavalent chromium

All that stuff is essentially banned under ROHS.

https://www.rohsguide.com/rohs-substances.htm

[r080]

I definitely missed the star washers under the rivets. That makes some sense, but, per willienillie's point, you can't just tighten up those rivets. I would assume that star washer is harder than the rivet, and over time the whole connection could loosen.

[stallik]

Rivets with star washers, bolts with star washers. Either may loosen over time but I'm sure that if Pete is sticking with the rivets, it's down to him and his customers being satisfied with the low failure rate.

Maybe. I just like the look of his work

[davent]

I'd think you'd be able to tighten rivets with a chasing hammer or other small peened hammer.
dave

[Rob Strand]

Rivets with star washers, bolts with star washers. Either may loosen over time but I'm sure that if Pete is sticking with the rivets, it's down to him and his customers being satisfied with the low failure rate.

An isolated rivet like that with no load doesn't have much reason to come loose.   It's not like a rivet on a ladder which is copping a beating on a daily basis.