pre amp noise

[dscottguitars]

I am having a hard time with noise in my preamp's distortion channel.  There is buzz when it's fully cranked.  If I turn down the gain a bit it's fine but then the distortion is dropped considerably.  

The schematic drawing shows only channel 2 and the power amp section.  There is more to the amp than that but it has nothing to do with any problems I'm having.  The preamp tubes are full dual triode, wired in parallel.  I thought this would help with noise.  The values of the components are: double for the resistors and half for the capacitors compared to using a single triode.

I have the clean channel working great and quiet.  If I disconnect the preamp and run just the power amp alone it is very quiet.  I can hook up a presonus tube preamp to it then and I get the great distortion I am looking for.

So that makes me wonder about the power to the preamp and why having a separate one is quiet and one built in is buzzy?

And, since the clean channel seems to be quiet enough, is there something inherent about high gain for distortion and noise?

Thanks...

Daniel

[Willabe]

Could be lead dress or grounding? The more gain the more you have to be carefull with lead dress and grounding.

Can you post a few good pics?


            Brad     

[dscottguitars]

I have addressed both of those issues.  Everything is grounded at one place in the preamp with a wire from that spot to the chassis.  I have the input jack grounded at the chassis next to the jack now, where I had it going to the one star spot, but now change.  No loops that I can see.  A wire then goes goes to the power amp at the center tap ground spot.    I have a two star ground on the power amp, one for the center tap, screen filter, bias, and cathode.  The other is for the PI, master, and jacks.  I've tried grounding the preamp to either of them, but no difference.

The preamp is a separate unit that I bolt on the bottom of the power amp.  Moving it around makes no difference in the noise.  I will try to take some pics and post them.  I have completely disconnected the first channel and the relay switch hoping there was a lead dress problem there, but no change.  I use shielded wires and the only crossing I may have was with channel one's wires going to the relay, but since I have disconnected it, I don't think it is the problem.

I have not tried using shielded wires for the boost going to the jacks on the footswitch.  They are longer, about 10".  Could that be it?

What really puzzles me is that channel one is pretty quiet, only a slight buzz and that goes away when I turn the feedback pot down, but the distortion channel is too loud.  I have checked other amps like Fender and Marshall 100 both watt and their lead channel is much quieter. 

I have also tried a small cap, 470pf and 0.001uf from the output pins to the cathode on both tubes but that didn't do anything.

I am going to hook everything back up, take pics and post them,

Thanks for your response...

Daniel

[Willabe]

Hi Daniel,

I have addressed both of those issues.  Everything is grounded at one place in the preamp with a wire from that spot to the chassis.  I have the input jack grounded at the chassis next to the jack now, where I had it going to the one star spot, but now change.  No loops that I can see. 

The chassis can form a loop. Here's a good link if you haven't read it yet on grounding.

http://www.valvewizard.co.uk/Grounding.html

A wire then goes goes to the power amp at the center tap ground spot. I have a two star ground on the power amp, one for the center tap, screen filter, bias, and cathode.  The other is for the PI, master, and jacks.  I've tried grounding the preamp to either of them, but no difference.

The PT CT and the reservoir (1'st B+ filter cap/s) caps ground has the most ground current in the whole amp and therefore is the noisest ground.

I have not tried using shielded wires for the boost going to the jacks on the footswitch.  They are longer, about 10".  Could that be it?

I can't tell from your schemo what/where the boost is but if it is a grid wire then yes it could be the source.

What really puzzles me is that channel one is pretty quiet, only a slight buzz and that goes away when I turn the feedback pot down, but the distortion channel is too loud.

Try disconecting the NFB loop and see if the noise goes away.


            Brad     

   

[HotBluePlates]

Where's the rest of the amp? Your schematic shows the very beginning of the preamp and the phase inverter/output stage.

[dscottguitars]

I have addressed both of those issues.  Everything is grounded at one place in the preamp with a wire from that spot to the chassis.  I have the input jack grounded at the chassis next to the jack now, where I had it going to the one star spot, but now change.  No loops that I can see.

The chassis can form a loop. Here's a good link if you haven't read it yet on grounding.

http://www.valvewizard.co.uk/Grounding.html

I read that and think I do what is required.  There is a part about putting the cathode section through a shield also, I may try that.

I have not tried using shielded wires for the boost going to the jacks on the footswitch.  They are longer, about 10".  Could that be it?

I can't tell from your schemo what/where the boost is but if it is a grid wire then yes it could be the source.

They are the cathode ground connections for the caps with the FW on them.

What really puzzles me is that channel one is pretty quiet, only a slight buzz and that goes away when I turn the feedback pot down, but the distortion channel is too loud.

Try disconecting the NFB loop and see if the noise goes away.


            Brad     

   

Are you sure it's okay to disconnect the NFB from the speaker tap?

[dscottguitars]

Where's the rest of the amp? Your schematic shows the very beginning of the preamp and the phase inverter/output stage.

I mentioned that already, it's not part of the problem.  I have isolated it to channel 2 only.  But I will post the entire schematic.

[dscottguitars]

Pic 1: front of the amp, upside down
Pic 2: channel 1
Pic 3: channel 2

[dscottguitars]

Pic 4: under channel 2
Pic 5: under channel 1
Pic 6: one side of power amp
Pic 7: other side of power amp

[Willabe]

Are you sure it's okay to disconnect the NFB from the speaker tap?

Yes.


         Brad     

[dscottguitars]

Hey Brad, that did something!  The noise is much much less, but the volume is too, although the distortion is still there.  But now I am confused because using the NFB and cutting it down with the pot would make it louder and increase the buzz.  I use this to get more output tube distortion, with the breakup point being more at lower volume using little NFB.  But with no NFB the buzz got much quieter.  So, the buzz got quieter with more NFB or none and a little made it buzz.  What gives??

[dscottguitars]

I just removed the two 820K resistors on the volume pot that I used to decrease the gain and the noise is back with the volume up to what I expect.  So, I'm not sure what is going on.

[Willabe]

What gives??

Dan that's a good question.      

On PP amps with a NFB loop if the OT primary wires are reversed, ie, start/finish, the amp can have all kinds of sound/noise problems. From full wail to odd noises at different control settings.

Try reversing the OT primary plate leads from tube to tube. Then reconnect the NFB loop. Still good? Volume back? Or is it screaming high pitch feedback?

I'm not sure about the OT plate leads because your output volume dropped.

HBP or 1 of the other more experanced guys will have better input on this.

From the pics you posted of the preamp I think it's gonna be hard to chase down any noise problems. It's very tightly packed in there.     That's not good for an amp with high gain. Why did you use such a small chassis for the preamp?


               Brad      

[Willabe]

I just removed the two 820K resistors on the volume pot that I used to decrease the gain and the noise is back with the volume up to what I expect.  So, I'm not sure what is going on.

You have quite a few connections that are floating and not soldered to eyelets or turrets. You could have a number of bad/weak or cold solder joints also.


             Brad     

[dscottguitars]

From the pics you posted of the preamp I think it's gonna be hard to chase down any noise problems. It's very tightly packed in there.     That's not good for an amp with high gain. Why did you use such a small chassis for the preamp?


               Brad      

It is all I had around that would fit and could be add to the power amp without having a separate preamp.  I'm thinking I may build a separate preamp instead.  I see some power transformers for only $20 or so that would work.  Which leads me to another question:

Could one use two transformers in one amp, one for the power amp and a small one for the preamp?  I notice that plugging in pedals, preamps etc. to an amp does not have any increased gain noise but my goal in making an amp is to not need pedals for distortion, only 2 channels and all in one unit.

[Willabe]

Could one use two transformers in one amp, one for the power amp and a small one for the preamp? but my goal in making an amp is to not need pedals for distortion, only 2 channels and all in one unit.

Yes, but it's a waste of space and money.

Once you go to a 2 channel (separately voiced) high gain amp compared to a 2 input/2 channel Fender/Marshall type amp you need to think it through more. There's things you can get away with in a lower gain amp that you can't get away with in a high gain amp as far as layout, lead dress and space.


            Brad     

[Willabe]

Here Dan go in to this link and there's a lot of pictures of well layout and well executed lead dress amps to study. There built by fellow forum members. It can be found at the top of the web page, click on media for future referance. 

http://www.el34world.com/Forum/index.php?action=media


            Brad     

[sluckey]

I see a few problems with your schematic...

1. V2-A grid has no dc path to the cathode. Biasing cannot be established.

2. V3 grid also has no dc path to the cathode. Biasing cannot be established.

3. V2-A has no coupling cap on the plate. DC on volume pot.

4. 4700µF cap and regulator chip reference pin need to be connected to negative bridge terminal.

[dscottguitars]

Thanks Steve...

Parts 3. and 4.  were just mistakes in the drawing. 

But I don't understand the first two:  DC path to cathode.  What do you mean?

Here is the corrected drawing...

[HotBluePlates]

He means your 6.8k mid resistor is on the wrong end of the caps, so there's no d.c. path from the grid to ground. It's blocked by the 0.047uF cap.

So buzz might be poor grounding or ground path from a filter cap to the cathode(s) of the tube(s) it feeds. Or it might be a poor solder joint, or a wire routing issue.

Respectfully speaking, you've put 10lbs of sh!t in a 5lb bag by using the small chassis for each half of your amp. And the rack mounting style may be causing a problem. Do you normally have the amp bolted together as shown in this first pic?

So in your first picture of the total amp, is the 2nd channel (the noisy one) on the right side of the input jack?

I asked the last 2 questions because if the 2nd channel is to the right (as see from the front panel with the open sides up), and the preamp is adjacent to the power amp as shown in your first set of pics, you have the 2nd channel over the power supply/rectifier area.

That almost guarantees noise will make its way into low-level preamp circuitry, and with following amplification you will hear noise of some kind. That assumes your wiring is perfect throughout.

I think you're driving yourself crazy with the folded-over layout, unconventional mounting of components in the preamp and the copious use of shielded wire. A good layout mostly doesn't need that shielded wire, and I go cross-eyed trying to figure out what's where. Must be a bear for you to get a soldering iron in there to take care of things.

This might deserve a different layout and a bigger chassis. Do you need rack-mounting capability? We might be able to help you with a better layout approach if you define the want-to-haves and the must-haves.

[sluckey]

But I don't understand the first two:  DC path to cathode.  What do you mean?

Let me say it another way... There is no grid return/leak resistor to ground for V2 or V3. Start at the grid and try to follow the dc path to ground. You can't. The path is blocked by the three TS caps. The grid needs this dc path which also leads to the cathode in order to establish bias. Without that path to establish proper bias the tube will be in some crazy undetermined state.

[dscottguitars]

AAAAHHHHHH!  Another mistake!  Both of the 6.8K resistors on the tone circuit are mounted off of the pot, not the caps.  Sorry.

And I know I've got a lot in those two chassis.  I will start over with a large proper chassis.

It is definitely hard to solder and unsolder the mistakes, trials and such as you can see the melted marks on the wires.  I should have played the game 'Operation' more when I was a kid.  LOL

Channel two is over the left side looking at it from the front on that pic.  It's not over the main power supply.  I can lift and move the preamp some and there's no change in any buzz or noise.

[dscottguitars]

Thanks everyone for your help and suggestions.  I see that I need to remake this in a proper large chassis with room enough for everything.  Hopefully when I'm finished I won't have to post here about it...

Daniel

[HotBluePlates]

AAAAHHHHHH!  Another mistake!  Both of the 6.8K resistors on the tone circuit are mounted off of the pot, not the caps.  Sorry.

No. Your "mistake" in the physical amp wired the tone stack properly.

You had the drawing wrong, which you'll see if you compare your tone circuit to the Deluxe Reverb schematic.

Channel two is over the left side looking at it from the front on that pic.  It's not over the main power supply.  I can lift and move the preamp some and there's no change in any buzz or noise.

So the quiet side, channel 1, is over the power supply. Then noise from the power supply induced in the grid circuits of channel 2 is not the cause.

Did I see a filter cap for the preamp over on channel 1's side then? Is it possible you don't have a good, solid ground to channel 2's side from that cap?

And why the arrangement of the filter cap nodes for the preamp? You diagram shows node D is used for the shared input stage and channel 1's second stage, but node E is arranged after node D, but only used for channel 2's second stage. I'd have thought you'd use node D for the 2nd stage of both channels and node E for the shared input stage.

If you grounded through the chassis for the node E filter cap only used by the 2nd channel, maybe that's a poor ground or ground current for it is mixing with something else providing interference.

[dscottguitars]

Yes, I had the drawing wrong, but the preamp is connected correctly.

There are two filter caps under channel one.  The blue one, a 20uF and a black one, a 10uF.  They and all the grounds go to the same lug.  From that lug, two wires: to the ground in the main power amp and another to the preamp chassis.  The ground solder is good and solid.

The input tube, V1 and channel 1 tube, V2 is shared by node D with the 20uF cap.  Channel 2 tube, V3 is on node E.  The reason for this is voltage.  I have more voltage on D and less on E.  I did that to have high headroom in channel one and high gain on channel two with lower voltage: ~160v on that plate vs. 210 on the other plates of V1 and V2.

I am ordering a board off the parts page and will drill my own holes, put eyelets in and try to make a nicer layout for all the components.  I have another project: 2 6L6's, but same preamp setup.  It's a finished model and I was trying to get this one discussed done and 'perfected' before I went on to the new one.  I have all the parts for it but it will take some planning to get it laid out nice.  See the pics of the chassis parts...

[guitardude57]

Holy mackerel... that thing is layered like a lasagna.  I am surprised there is nothing but noise.
I think once you find the spots that are mis-hung and unsoldered, it may be useable.
A simple circuit like a spitfire can be built in a matchbox, almost.
Do yourself a favor, try to go with as large a chassis you can muster, with multiple channel/high gain amps.
Sure is worth the extra real estate.  And less struggle.

[Willabe]

Hopefully when I'm finished I won't have to post here about it...

Yes you do and with pictures!


           Brad     

[dscottguitars]

Hey Brad,

Here is a pic of the guts (so far).  I'm almost finished, just waiting for the bias caps and isolated jacks-Marshall style...

This is my first turret board, and my layout wasn't perfect.  I've drawn up another one, with better planning, to use for the next amp.

Any suggestions?

Daniel

[Willabe]

Daniel, that is soooo much better of a build than the last 1! Totally night and day.    

I still say it's very dangerous to run transformer fly leads/wires through a hole in the chassis without rubber grommets. Highest voltages and current in the amp run through those wires.

Let us know how it sounds when you finish it.


               Brad      

[dscottguitars]

Thanks...

Those holes are large- 1.5" in diameter, they have been ground with no sharp edges and I'm not sure they make grommets that big.  The wires don't really rub/touch the metal at all and with it being grounded to the mains, I'm sure it is safe.

My next build will have smaller holes with grommets though.

[HotBluePlates]

I'm not sure it was covered in the posts... Did you ever swap OT primary wires to change your NFB loop from positive feedback (noise before removal, and increasing noise with "increasing negative feedback") to the desired negative feedback?

[dscottguitars]

No I didn't.  But if I understand you correctly, it wouldn't have cured the issue because channel one was working flawlessly with no buzz and great volume.  It was only in channel two, with distortion that the buzz was present.  Changing the OT wires I believe would have cause a loud uncontrollable howl or squeal.  I've had that before.  I'm hoping in this new build I'll have good results with better grounding, better separation of components and I'm isolating the jacks from the chassis.

The preamp, channel two -aka-distortion channel, is the only thing thing that made too much noise.  I took out the preamp  and have it used just as a power amp with a volume control, phase inverter and the four 6L6's; it is as quiet as a church mouse now.  It's not really loud enough with just my guitar but if I plug in my presonus preamp it sounds great.

[PRR]

One thing about a relay: you can mount it *near* the stuff it is switching, avoid those looooong signal leads spidering all over the chassis, inviting buzz and oscillation. Perhaps right behind the pots?

[dscottguitars]

If you mean the white wires, they are high quality aircraft shielded wire.  I could possibly move only the relay switch in the space by the input jack.  But I thought the shielded wires would be fine.

Also, is it okay to have long lines going to the jack for the switching?  I put it where it is to shorten those...

[dscottguitars]

Well here are the guts finished!  And, it sounds great.  Everything works nicely, very quiet with only hiss which is what I expect from the NOS Allen Bradley carbon comp resistors used.

The only thing I can't figure out is the scratchy noise on the negative feedback pot.  In my other builds I could turn the volume down and the NFB knob would be much less scratchy, but not with this one.  I posted a new topic on that subject too.

And, I think I may have found out some of the buzz issues with the spiders nest of an amp earlier in this post. Using the first input tube's cathode bypass cap as a footswitch boost does not work!  I had the same buzz issues with this one and grounded that at the filter cap and it went away.  I will be trying that on the other one now...

Next is a nice walnut case to house this unit but my workbench is my deck and it's full of snow.  So it may be a bit before I get that going...

Thanks for all your help and suggestions,

Daniel