Welcome to Solid State Guitar Amp Forum | DIY Guitar Amplifiers. Please login or sign up.

April 23, 2024, 05:34:22 PM

Login with username, password and session length

Recent Posts

 

Rod Elliott's Project 27 and Amp

Started by ghoshsubha444@gmail.com, February 27, 2013, 04:54:55 PM

Previous topic - Next topic

ghoshsubha444@gmail.com

Quote from: newbiediy on March 11, 2013, 10:51:14 AM
Just sharing my newbish "experience".
I also had noise problem when first building my own amp.
But it was BEFORE I put it in a chassis. After putting the whole thing into a chassis and connecting input ground to chassis, it was noiseless.
As far as I know, you should connect only ONE ground point to chassis. Please correct me if I'm wrong.

which amp you built?

newbiediy

It was Little Rebel. Only 2W amp, so I didn't expect much noise from it.
But surprisingly it was so noisy before mounting it in a chassis. And surprisingly quiet after chassis work. :)
Sorry for my bad English. We say "laik dis" instead of "like this" in Facebook. :P

ghoshsubha444@gmail.com

#32
Quote from: Roly on March 11, 2013, 11:01:12 AM
Overall layout

Apart from a grounded full metal chassis you will notice that Rod has added a shield over the preamp board (which may be metal such as aluminium, or single-sided PCB laminate, but must also be grounded); and another shield around the input sockets.

Preamp


Power amp


I'd just like to reinforce the point that Rod Elliott is a very professional and capable designer and frankly a cut or two above the vast majority of people who post circuits to the web. I've had 40-something years in electronics design but I would be very careful about second-guessing Rod or trying to "improve" one of his designs.


Now part of your problem seems to be trying to do too much at once.  Particularly with a new build we try to take each stage a step at a time, identify and correct any problems with that stage before moving on to the next, and in audio amps this generally means starting with the power supply, then the power amp, and so on moving stage by stage back towards the input.

The difference between having boards scattered over the bench, and mounted in a metal chassis, can be very large where picking up external electrical hum and noise is concerned.  The grounded metal chassis makes a very effective electrostatic screen around the assembly.

Now this particular amp design is a little unusual in that it uses some current feedback, and Rod has some particular things to say about how the main amp section should be laid out and wired to avoid instability.  I suggest you read them again closely and mount your power supply, main amp, and output sockets in your chassis then test at that stage that it is working properly.  You can feed a test signal from an MP3 player or similar into the main amp input, and you should be able to get a clean and loud output from your speaker (which must also be properly mounted in its enclosure).

Once you have got to that point you can then add the next stage in front of that and test again, then the next and so on until you finally get to the input sockets.

Before you get too deeply into distributing DC power around your chassis you need to read this thread;

http://www.ssguitar.com/index.php?topic=288.0

Poor distribution and incorrect grounding causes a lot of preventable problems with amp builds, so it's important to have a single chassis signal ground point, and to distribute your power and ground lines radially from there, not daisy-chain fashion as your assembly drawing shows.

It is also important with single point grounding that you use sockets that are fully insulated from the chassis.


Q.1 I suspect that grounding may be the cause of the problem.

You appear to have used hookup wire for the signal interconnects between boards.  Generally speaking it is better to use screened cable for signal interconnects with the screen only connected at the receiving end, the common grounds going around via the single ground point.

Q.3 a line output can be added later using a simple op-amp buffer taking its signal from the input of the main amp.

Q.4 Not really, three pin regulators generally have better noise performance than zeners, but in any case supply lines should be well bypassed with both large value electrolytic caps and some smaller ones such as 0.1uF close to each IC, and particularly right at the input and output of three-pin regulators.

Q.5 Just about any reasonably high gain audio transistor.  The BC549 is a member of  very old family that go back to the BC109, and the main difference in the BCxx9 family are the packages and lead order.  The suffixes A, B and C are the sorted gain ranges being "lowish", "medium" and "high to very high".  Similarly the original BCxx7, BCxx8, and BCxx9 where 'rough", "smooth", and "very smooth" in terms of noise and audio quality, however today even the BCxx9 family are surpassed by audio-specific op-amps that are better in every respect.

The main thing when looking at the data sheet for a possible substitute is that it will withstand the supply voltage and pass the required current, with a safety margin for both, then that it has a gain, hfe, the same or higher.  Then you chose the one with the lowest noise figure, but with modern transistors the noise figure isn't generally an issue except in really demanding circuits such as microphone preamps.


So, start getting the initial parts into a chassis, mains wiring, transformer, power supply, and main amp, and testing each stage as it is completed, then as you add stages in front of the main amp we can deal with the problems as they arise.

HTH

Thoroughly Explained :) :) Thanks sir.
I must go along step by step and first thing first.....

I have already completed power supply,power amp and p27 (apart from some pedals).

Let me prepare some documents about the schematics and the layouts to verify here..

Let me tell you my desire. all that  I want is to build some beautiful tone. may be i have only clean sound! But that must be beautiful so that when playing i have feeling for it. And i have realized that incremental development is the only way to survive, otherwise you will be frustrated and soon will be out of DIY stuffs.

will be back soon....

ghoshsubha444@gmail.com

Quote from: newbiediy on March 11, 2013, 11:24:32 AM
It was Little Rebel. Only 2W amp, so I didn't expect much noise from it.
But surprisingly it was so noisy before mounting it in a chassis. And surprisingly quiet after chassis work. :)

:dbtu: :dbtu: :dbtu: :dbtu: :dbtu: :dbtu:

J M Fahey

FWIW, here's one of my SS (obviously ;)) High Gain Guitar Amps.

1) first from outside, to know what we are talking about:



2) then the inside, showing PCB mounting, compact and simple wiring, internal Aluminum shields:





And yes, it does have an output transformer  ;)

ghoshsubha444@gmail.com

#35
Quote from: J M Fahey on March 11, 2013, 06:52:29 PM
FWIW, here's one of my SS (obviously ;)) High Gain Guitar Amps.

1) first from outside, to know what we are talking about:



2) then the inside, showing PCB mounting, compact and simple wiring, internal Aluminum shields:





And yes, it does have an output transformer  ;)

Do you run production?  :) :)

Did you paint on the transformer??

ghoshsubha444@gmail.com

This is the main regulator circuit i built.

Additionally it contains 1 lm317 for taking 9v output.

Please verify whether my idea is clear on 'grounding'. please ignore my badpainting :o

Roly

Nice one JM  :dbtu:  The kind of elegant simplicity it takes years to achieve.   ;)


Quote from: ghoshsubha444all that I want is to build some beautiful tone

Oh, is that all?  The hard stuff first eh?   :lmao:

Well first we get it all working, and then we take Berlin, okay?.


The power supply circuit looks fine, but the 0.1uF and 1r/2W in series aren't needed.

Yes, the grounds should distribute radially from a single point, so you've got that idea.  This also applies to the distribution of power on the +ve side.  Each section should have one ground connection, and only one, or you will get hum from the dreaded "earth loop".

The single ground point is normally a large bolt through the chassis with nut and lockwasher, then washer, the mains earth, washer & nut, washer then the other earths with another washer and nut.  It ends up being a bit of a tree.  It is very important to make sure the mains earth is very firmly attached, and common practice is to crimp a closed ring terminal on the end on the earth wire so even if the nuts should work loose it will hopefully still be making contact.  If you can't get such a terminal you can fake this by looping the earth wire around and twisting it around itself to form a closed loop around the bolt.

Mains earthing is vitally important with guitar amps because the chassis common is carried through to the guitar strings, and if it should become live, so do you (and you may not be live too much longer if that happens!).  There should also be a suitable fuse in a suitable mains-rated holder in the mains infeed as fire protection.
If you say theory and practice don't agree you haven't applied enough theory.

ghoshsubha444@gmail.com

Quote from: Roly on March 12, 2013, 04:55:46 AM

Well first we get it all working, and then we take Berlin, okay?.

ok :) :)

Quote from: Roly on March 12, 2013, 04:55:46 AM

Yes, the grounds should distribute radially from a single point, so you've got that idea.  This also applies to the distribution of power on the +ve side.  Each section should have one ground connection, and only one, or you will get hum from the dreaded "earth loop".

I am trying to refine my image and understanding. Please check my attachments.

Quote from: Roly on March 12, 2013, 04:55:46 AM
The single ground point is normally a large bolt through the chassis with nut and lockwasher, then washer, the mains earth, washer & nut, washer then the other earths with another washer and nut. 

What is 'other earth'?

is Bus connection of +Ve // -Ve permissible?

phatt

#39
Re Other Earth Q,
Some folks get confused with Earth, Ground and Common as these terms can all be referring to the same Node in a circuit,,,, but Not Always.

Not sure how your circuit is wired now but this pic might help.

If you look closely at Fahey Amp inside pic shows the mains EARTH wire bolted to chassis and then a white wire (Circuit Common) traveling from there to the centre of those two big filter caps of the power supply.

And a special thanks to Jaun for showing the inside pics. :dbtu:
And I like the white front panel,, something you can actually read in the dark and knobs that actually point.
Phil.

J M Fahey

Thanks  :-[
And yes, the picture clearly shows the back panel main grounding point, which I pop rivet, so it's impossible to unbolt by mistake .
Eagle eyed Phil  8| correctly found that the green/yellow ground wire goes from the center of the IEC connector to ground, and a white wire goes to the center of the main filters which is the "Electrical ground".
All other Amp grounds are referred to that.
In fact, if you look closely, I also send the power amp input ground to the electrical ground with a separate white wire, I wanted a good straight path and a PCB track would have zigzagged more than what I wanted.
Also speaker ground returns straight to main electrical ground.

joecool85

Quote from: J M Fahey on March 12, 2013, 12:34:04 PM
Thanks  :-[
And yes, the picture clearly shows the back panel main grounding point, which I pop rivet, so it's impossible to unbolt by mistake .
Eagle eyed Phil  8| correctly found that the green/yellow ground wire goes from the center of the IEC connector to ground, and a white wire goes to the center of the main filters which is the "Electrical ground".
All other Amp grounds are referred to that.
In fact, if you look closely, I also send the power amp input ground to the electrical ground with a separate white wire, I wanted a good straight path and a PCB track would have zigzagged more than what I wanted.
Also speaker ground returns straight to main electrical ground.

I still want to buy one of your amps, Juan.  It doesn't seem commercial amps sold in the US are built to this level of quality.
Life is what you make it.
Still rockin' the Dean Markley K-20X
thatraymond.com

J M Fahey

OK, thanks.
We'll arrange something.
JM

phatt

Watch out Joe!!
If you notice an old dusty Argentinian Falcon Car in your street with a few bullet holes in the panels and being driven by a mysterious looking man wearing dark sunglasses and asking for Joe's place.

Invite him in,,,Use the password *SSguitar* and ask if there is a Fahey amplifier in the Boot,,,Underneath the bones of some long forgotten politician. :police:

Sorry Jaun,,, I could not resist the temptation ;D ;D

Phil.

phatt

@ ghoshsubha444,
I'm fairly certain that Rods pages explains some of the loop holes in layout on at least one of those Many pages.

You may just need to go a little deeper into that site,, yes it is vast and you can spend hours finding hidden gems of info on His site,, but maybe worthwhile.    ;).

Re the linking of power rails to many boards,,,
Depends on the setup,,,, so very hard to know as the setup you describe seems rather complex. (Headscratch?)

i personally would not want to run a circuit from split rails then run single rail and back to dual rail.
Same  reason why I avoid using pedals in the efx loops of Amplifiers,,, it just complicates an already complex issue.

I run all pedal type circuits *Before* main Amp circuit so no efx loops.
I do use a *Parrallel* efx loop which overcomes some of the issues when needed.

Phil.