Larry's Grounding Scheme on a 2204 circuit

[syscokid]

On the info that I have gathered so far, it seems that the PI filter cap and the Screens filter cap are to be grounded at seperate points from each other. But on the 2204, the PI and the Screens is shared with one multi-filter cap...right? Right now, I have the Mains cap and the PI/Screens cap, with the HT CT, grounded at the same point at one of the Power Transformers bolts ( I haven't tried to ground the Output Tubes here, yet ).

Also, the Master volume: Since it's a control that goes between the Tone Stack and PI, I grounded the MV pot, with the Tone Stack/Presense/Output Jack grounds, at one of the Output Transformers bolts.

AC cord is grounded right at one of the IEC's bolts.

Bias circuit is grounded by itself at a different PT bolt.

Pre-Amp volume, V2 cathode, & Pre-Amp filter cap are grounded at a solo bolt, between the Pre volume pot and the board.

Input Jack, V1 cathode, & Heater CT are grounded to a solo bolt, just to left of the board ( V1 area ).

All filter caps were properly formed.

Overall, the amp is much quieter from it's initial stage, but it's still far from being "dead quiet" as others have described. I still get a little bit of hum with the MV dimed, while the Pre-Amp is on zero and guitar plugged into High. And of coarse, more noise comes in as you crank the Pre-Amp. I wouldn't consider the noise or hum excessive, but there seems to be varying opinion on what someone would consider as being "quiet".

So, is this as good as it's going to get?

Am I on the right track here?

Suggestions and comments always appreciated...

[joey]

If you are looking for dead quiet, avoid chassis grounding schemes at all costs. I typically ground all circuitry straight to the filter cap return of the stage they feed, and besides the IEC safety ground make one, and only one connection to the Chassis for the reference return (circuit ground) right at the input.

[syscokid]

..... make one, and only one connection to the Chassis for the reference return (circuit ground) right at the input.

Hi Joey,

I'm totally confused, now. Can you explain this differently ...for a Dumb-Ass Truck Driver like me?

[novosibir]

By principle and theoretically joey is right - the best grounding scheme is:
- each stage does have its own filter cap
- all grounds of every stage are connected to their corresponding filter cap negative (sub star)
- all sub stars are routed together in the order, as the circuit is running (ground bus) - still isolated from the chassis (!)
- the ground bus is connected to the chassis at only one point

That's the way, how I'm doing the grounding in my own amps.

But in Marshalls, Fenders, aso. and in their derivates & clones not seldom are some (or more) designer bugs - and sometimes you may think, that the amp designers better should have been selling burgers at McD, instead of designing tube amps.

But anyway, there's a way, not to make all the efforts of a chassis free ground w/ a separate ground bus in a marshall-esqe amp, to get it (almost) dead quiet. The way is, to USE the chassis instead of a ground bus as a ground plane. The different 'sub ground spots' have to be well choosen and proven by empiric, what I've done and refined over years in a lot of different Marshall amps, which I've modded, upgraded or just serviced - and that's, what now is known as the "Larry's grounding scheme".

To get the idea, how my grounding scheme is done, first read some:

http://forum.metroamp.com/viewtopic.php ... 25&start=0" onclick="window.open(this.href);return false;

http://forum.metroamp.com/viewtopic.php ... 1&p=335678" onclick="window.open(this.href);return false;

http://forum.metroamp.com/viewtopic.php?f=15&t=27641" onclick="window.open(this.href);return false;

http://forum.metroamp.com/viewtopic.php?f=4&t=25351" onclick="window.open(this.href);return false;

And some additional hints:

NEVER use a PT bolt as a ground spot!
ALWAYS run the OT's secondary common to the speaker jack! Don't ground it anywhere to the chassis!
Run a regular (22 AWG) cable from the speaker jack's shield to the spot, where the PI is grounded.
Use lock washers between the ground lugs and the chassis, to enable good contact.

One of the designers bug mentioned beyond is the double filter in the 2204 for screens & PI w/ common ground.
Here I'd recommend to give this filter (one or both halfes - try it, what you like better!) to the screens and I'd mount an additional filter only for the PI, what then will be grounded to ground #3 of my grounding scheme.

Larry

[ryu]

when you say "all grounds of every stage are connected to their corresponding filter cap negative"
for ex v1a and b should go to the red circled point in the layout?
then we disconnect the ground point of the dual can cap, the same for all and just finally "ground" everything at the same point?
http://img600.imageshack.us/img600/2441/2203y.jpg" onclick="window.open(this.href);return false;

[joey]

By principle and theoretically joey is right - the best grounding scheme is:
- each stage does have its own filter cap

I don't know if I would go that farsuch as making it a rule per se, although I do this on some of mine mostly because I like the Tone, and response and it makes the layout a little easier. So long as two antiphase stages share a node (cathode followers are excluded from this rule) , and everything is grounded with in close enough proximity to the return node it is fine.

But anyway, there's a way, not to make all the efforts of a chassis free ground w/ a separate ground bus in a marshall-esqe amp, to get it (almost) dead quiet. The way is, to USE the chassis instead of a ground bus as a ground plane. The different 'sub ground spots' have to be well choosen and proven by empiric, what I've done and refined over years in a lot of different Marshall amps, which I've modded, upgraded or just serviced - and that's, what now is known as the "Larry's grounding scheme".

the only thing I slightly disagree with is that true ground planes function quite differently. I myself couldn't see the difference, at first, but the more I learned about them the more I changed my view. I do see what you are saying, and that you would be relying on a proximity effect to the return of each node to make the chasis scheme work as you would want to prohibit the higher current returns from finding a lower impedance return path outside its local loop, which is sort of the principal behind a plane, but I still think the two are quite different.

[joey]

when you say "all grounds of every stage are connected to their corresponding filter cap negative"
for ex v1a and b should go to the red circled point in the layout?
then we disconnect the ground point of the dual can cap, the same for all and just finally "ground" everything at the same point?
http://img600.imageshack.us/img600/2441/2203y.jpg" onclick="window.open(this.href);return false;

No, more like this:



The only exception really depends on the type of loop you use.

[syscokid]

Besides going thru those suggested threads, which led me to more threads, then GeoFex, then Aiken's tutorial, figuring what Joey posted and illustrated...I decided to have a few beers!

Just kidding!

I repositioned any grounds that were using the OT's or the PT's mounting bolts.

I moved the Heater CT to position 5 instead of position 1. I did this because I did not notice any difference and eliminated the long length of wire to jump that distance.

I tried a few other combinations, but this is what I settled on in the pic below(...for now). To make sure my ears weren't playing a trick on me: I used a sound level meter, positioned about 4" from the speaker. I set ther amp's Pre Amp volume at 8, MV at 8, Tones flat, Presense on zero. One test was no guitar plugged in: 78 to 85 db. The other test was with guitar plugged into the High input jack: 90 to 98 db's. Most of it was hum.

You know what works really good? ...........................How about making sure there's a good-working preamp tube in V1... Originally I tried two different tubes. What's the chance's that two relatively new tubes would be bad....someone shoot me...

95% of the hum is now gone: I'm now getting noise levels of less than 60 db's - no guitar; 72 db's with guitar, which is all about the hissing at these extreme volume levels, but turn the Master down to about 4 ( Pre on 8 ), and it becomes very quiet but still very loud...

IMG_1443 - Copy.JPG (164.37 KiB) Viewed 8596 times

[joey]

dude #1 rule of trouble shooting: always try the easy stuff first. I have been guilty of this too, you'd think I would learn LOL

[syscokid]

What was really funny, after I installed the current tube, I thought that this tube was bad because I could barely hear any noise coming out of the amp. It wasn't till I strumed the guitar that I realized what had happened...

And also, I did learn a few new things about grounding.

Now off to the next adventure: A "Vari-Watt" circuit from SkipzCircuits...

[Maria465]

Larry's grounding scheme on a 2204 circuit is meticulous and effective. His attention to detail ensures a robust electrical system, minimizing the risk of interference and enhancing overall safety. The thoughtful placement of grounding points demonstrates Larry's expertise in optimizing circuit performance. It's a well-executed plan that reflects a deep understanding of electrical engineering principles.Nrega Job Card List