Fiberboard Conductive?

[jeff]

  I recently rewired my amp and found something funny that I can't explain. After I rewired it the gain pot sounded scratchy, not like a dirty pot but like a presence pot so I figured there was DC on the pot. I checked and there was ~.2V so I figured it was a leaky cap. I desoldered the end of the cap going to a 470K resistor on the board going to the gain knob and tested. There was no voltage on the cap but there was voltage on the pot. I disconnected the pot from the board giong to the 470K and there was no voltage on the pot. I checked the voltage at the 470K resistor which was now connected to nothing but two eyelets on the board and found ~.4V. I then desoldered the 470K resistor and there was ~.4V on the eyelet. I checked for flash or a loose piece of wire on both sides of the board to see if something was making a connection but there was nothing. I can't figure this one out. The eyelet that's giving me the .4V is about 3/16ths away from the 250V B+ for the preamp tubes. Is this too close? I can't figure out how there's voltage on that eyelet. Like I said there is no flash or drips of solder or strand of wire connecting the two on either side of the board. I don't think I'm getting a false read on my volt meter because the pot sounds scratchy like there's DC on it. I measured with an ohm meter that goes to 20M and there is no connection. The only thing I can think of is that the eyelets are too close together and the board is slightly conductive. Is there a minimum distance to how close you should have the eyelets. Any opinions, this is driving me crazy. By the way the fiberboard is from Mojo.

[phsyconoodler]

That's not the problem.However,you may have a poor ground somewhere causing the issue.
  It would help seeing your build and what circuit it is.
Sometimes the tube in line with your gain will do that too.

[jeff]

I'm getting the .4VDC on an eyelet that's not connected to anything. The board is clean on both sides, no solder or wire or scraps between the two. It's wierd. I can't explain it. The only thing I can think of is that it's too close to the B+ and is somehow going through the board.

[sluckey]

Is the board material the thin paper board that Fender used in old amps? Conductive boards was not uncommon on the old Fenders, especially with closely spaced eyelets.

[phsyconoodler]

Not likely on a new board.Maybe an old board that's damp for years.
  I'm thinking you need to re-examine your circuit or look elsewhere.With a gain control  it hooks directly to the tube grid(pin 2) unless you have some other design.Did you try another tube yet?And it couples to the rest of the circuit with coupling caps a swell.What about a bad treble cap?

[jeff]

The board looks exactally like what was is my PA-100. Like a black, heavy cardboard.

It's not the circuit. I removed the conponents from that part of the board and I'm getting a voltage on an eyelet that has nothing soldered to it. It's reading .4V and is 3/16ths of a inch next to my 250V B+

 I'll try a test later on tonight. I'll put two eyelets on some extra board material spaced the same distance apart. I'll sandwich a blank board under it and screw the boards to a piece of sheet metal and ground that. Then I'll run a wire from my B+ to one of them and read the voltage on the other. If there's nothing it's not the board.

[JayB]

Measure the board around that eyelet and see if you got voltage.

[phsyconoodler]

Are you using a digital meter?If so,that reading may be erroneous.They can jump around and sometimes give false readings.Try an anolog meter.
  I have yet to come across a conductive fibreboard unless it's pretty damn old and has absorbed moisture.And that's a whole lot of fibreboard.
  

[sluckey]

Do a google search for "fender conductive fiber board". You may be suprised at what turns up.

[jeff]

from http://www.guitaramplifierblueprinting.com/Victoria1.html

You see, the fiberboard is not a perfect electrical insulator. When conditions are right, micro voltages can creep across the board. Normally this isn't a big deal--we're talking about MICRO voltages here, but they can occasionally disrupt normal amplifier operation. The most common manifestation of the voltage creep is the annoying ticking noise some people have noticed. It occurs when the B+ on the 100K plate resistors for the first preamp tube creep across the board to the 68K grid resistors feeding the bright channel. On a Bassman, the proximity of the B+ (at approx. 300V) on the 100K's is literally 3/16" away from the 68K input resistors. These resistors feed the grid of the first preamp tube---a very sensitive section of the amp. The voltage this grid sees at idle should be 0V. When the B+ voltage creeps across the board it puts approx. .1volts on this grid.

Wow, now I can go back to believing I'm sane!

[jeff]

Where can you get those?

[sluckey]

Doug sells everything to make your own boards, ie, fiberglass board, turrets, eyelets, etc.

http://www.turretboards.com/   sells reproduction fiberglass boards for many old amps.

[phsyconoodler]

Yes, I agree that they can BECOME conductive,but this is a brand new board from Mojo.
  I have built quite a few amps with those boards and not one has given me a problem like that yet.And there are a great many older Fender amps out there that do not have conductive boards.
  I'm not saying yours isn't conductive,just trying to get you to cover all the bases before you condemn it.3/16" is not that close anyway.
What circuit do you have?That might help shed some light on another possible solution.

[sluckey]

I totally missed that it was a new Mojo board.   

[stingray_65]

I've got a megohm meter I use for things like this.

They used to be called "Meggers" and had a hand cranked generator, but new ones like mine use a few transistors and a step up tranny.

Living here in the countryside when I drive around I look for a mast with beam antenna's.

I've bought LOTS of goodies from widows of old hams operators, like my megohm meter. Well actually got it from the son of a ham radio operator.

Ray

[jeff]

Here's the circuit.

Would radio shack perf board be OK?

[jeff]

So no known drawbacks to using RS perf board? I don't want to get suprized again. I got some left over from the first amps I built. Funny thing is I used perf board for this amp in the first place. I redesigned the preamp and instead of adding two eyelets, totally redid it with the fiberboard because I thought it was better. Should have left well enough alone. Unfortunently I think I threw the old one out.

[sluckey]

The RS perfboard is not very rugged but should be fine in a stationary environment. If I was going to totally replace a circuit board in a guitar amp, I'd use high quality, rugged material, such as 1/8th inch fiberglass with turrets or eyelets.

BTW, what amp is this? And do you have any problems other than measuring a very slight voltage on that one terminal? Do you have any symptoms that actually affect the sound of the amp?

[jeff]

Well my two main concerns are:

1)The gain pots are scratchy

2)Having this voltage on the pot means as I turn up the gain the voltage on the grid increases

I'm worried about the next tubes bias. If the grid has .4V and the cathode has 1V the tube sees it as having only .6V bias.

[EL34]

Notice that there is no fiberboard material on my web site for sale?

It may be new, but I still do not like the stuff.

If you use the glass board and lugs, your problems will be none, beer will flow in all the fountains and you will receive gifts from above 

see this page for the board and lugs
http://www.hoffmanamps.com/MyStore/catalog/parts12.htm

[jeff]

It just seems inconcivable to me that a circuit board material would be conductive! Wouldn't of thought of that in my wildest dreams.

Just wish I knew that before I spent $40 on 14 of them. I figured that'd be a lifetime supply. Oh well live and learn.

I'm re-doing the board with Radio Shack perf board and that should be that.

                       Thank you to everybody for all the help.
                                           Jeff

[EL34]

Well, I heard that the coloring that makes the boards black is carbon based.
Not sure if that is true or not, but if it's carbon based then it can flow current.

On old Fenders, there's an eyelet junction or two where a couple plate resistors come together and the voltage is very high
There's a screw near there that holds the boards down to the chassis

I used to be able to measure lots of voltage leaking through the board in that area.
If you lifted the screw out, the voltage would go way up because the ground connection was lifted.

Sometimes it was so bad that I had to lift the parts out of the eyelets and suspend them in the air to fix certain problems.

I have also had to separate the top board and bottom board and shove some insulating material in between them in extreme cases

[jojokeo]

Ya know, it's getting close to Halloween so I suspect you are measuring ghost voltages!

Seriously, all you need to do is put a slight load there and if it's actually a voltage you'll still measure it, if not it will go away. This is common when working w/ DMMs and AC voltages. Atleast you can verify it or not?

[phsyconoodler]

I still don't think that's your problem.I'm betting it's a bad cap if you are using treble,bass,mid caps in your build.
  WHAT CIRCUIT ARE YOU USING?

[jeff]

I don't understand what you want from me.

Do you want the schematic of the whole amp?

My problem is a voltage on an eyelet that has nothing connected to it. When I wire components to it, it puts a voltage on the pot.

...the gain pot sounded scratchy, not like a dirty pot but like a presence pot so I figured there was DC on the pot. I checked and there was ~.2VDC so I figured it was a leaky cap. I desoldered the end of the cap going to a 470K resistor on the board going to the gain knob and tested. There was no voltage on the cap but there was voltage on the pot. I disconnected the pot from the board going to the 470K and there was no voltage on the pot. I checked the voltage at the 470K resistor which was now connected to nothing but two eyelets on the board and found ~.4VDC. I then desoldered the 470K resistor and there was ~.4VDC on the eyelet.

I tried to attach a picture again. Did it come up?

I think the voltage was lower in the first two pictures when the pot was connected but I accedentally wrote .4V for all

[jeff]

Seriously, all you need to do is put a slight load there and if it's actually a voltage you'll still measure it, if not it will go away. This is common when working w/ DMMs and AC voltages. Atleast you can verify it or not?

What did you have in mind for a load? I know if I wire from the eyelet to a 470K resistor to a pot to ground I can vary the voltage at the wiper by turning the knob. Does that help?

I rewired the amp using a different circuit board material and will check it out tomorrow.

[oldgitar335]

sirs,  kudos to all. could it possibly be the flux that melted and flowed around each eyelet, made the path between them ?
        all respects,     oldgitar   

[phsyconoodler]

what I meant was what circuit are you modelling the amp after?2204,Plexi,Dumble or your own design?
  The coupling caps,treble,mid and bass caps are what I'm interested in.If any of those are leaky it will show up in your pots.In your diagram there is a cap and then a resistor connected to one side of the plate resistor.That cap may be leaking.
   

[jeff]

ok, sorry I didn't get what you were asking.
The preamp is close to Doug's tube pedal project with a few tweaks. 2k7 instead of 2K2 cathode resistors bypassed by .68 caps. A second gain pot between V2 and V3 instead of resistors with no local feedback. Also there's a 470K resistor in series between the coupling caps and gain pots. Long tail 12AX7 phase inverter. Push pull 6V6 cathode biased power amp.

The picture I drew shows the first two tubes.
The scratchy pot in question is the first gain so I don't think it can be in the tone stack.

How can I test that cap? This is how I tried. I lifted the end off the circuit board that connects to the resistor to the gain pot and with the amp on was reading 0VDC on the cap.

The DC is on the pot when I totally remove the coupling cap so weather it's leaky or not is moot. Only when I break the conncetion from circuit board to the pot does it go away.

[jeff]

Look at this picture again.
In the first picture I'm getting a voltage on the other end of the coupling cap. So I too suspected a leaky cap.

In the second picture I lifted the cap from the circuit board and measured DC on the cap-0V. There was still voltage on the pot.

In the third picture I totally removed the cap and resistor going to the pot. There was no voltage on the pot but there was a voltage on the eyelet that has nothing connected to it.

[jojokeo]

Curious, what kind of grounding system are you using and how is that pot grounded specifically? I'm wondering if you're seeing something that's actually going on w/in your chassis? I would tend to agree w/ physco...seems unusual what you're experiencing w/ your pot and then you're also measuring a voltage w/ nothing on an eyelet - spooky amp there. These kind of things are very hard for us to troubleshoot when it's on a bench far far away. Hang in there and try not to get too frustrated, I know it's easier said than done sometimes. But, sooner or later you get a handle on it.

[jeff]

I think my grounds are good. Lug from pot soldered to back of pot. Pot attached to chassis.

I'm convinced the board itself is to blame. Maybe I got a bad batch. I've replaced it with Radio Shack perfboard. The layout of the eyelets is exactally the same. No problems with the scratchy gain knobs now. No voltages where they shouldn't be.

The only problem is it squeals like a piggy with the gain from 8-10. I may have moved the wires a bit when I put the new board in. I remember having to push them around to keep the amp happy when I built it. The only difference I can see is, since the perf board is raised up on spacers and the wires are not laying flat against the chassis. Would this have a bad effect?

[jojokeo]

I think my grounds are good. Lug from pot soldered to back of pot. Pot attached to chassis.

I don't support that myself. I don't like signal ground connecting to chassis ground that way. Also, you're asking for trouble relying on a mechanical connection but to each his own.

The only problem is it squeals like a piggy with the gain from 8-10. I may have moved the wires a bit when I put the new board in. I remember having to push them around to keep the amp happy when I built it. The only difference I can see is, since the perf board is raised up on spacers and the wires are not laying flat against the chassis. Would this have a bad effect?

Not usually. In all of my highest gain amps, I've never had this issue where the amp is so extrememly sensitive to wire placements. IMHO, good design, shielding, grounding, and lead dress goes a long way in preventing not only this but parasitics and other oddities as well.

Glad you got things under control and thanks for letting us know how far we can trust those boards too. These are my favorite kind of lessons that I don't have to experience for myself to learn from.

[jeff]

What's a better grounding system?

[PRR]

> seems inconcivable to me that a circuit board material would be conductive!

EVERYTHING conducts.

Fiberboard is usually perfectly acceptable around 400V 1Meg audio amplifiers. You won't see it around KiloVolt systems or GigaOhm electrometers.

There's many things it "could" be. But for now, why don't you "air-wire" that one joint (without touching the board) and see what it measures? Sure, a air-wire joint won't stand a long road-trip; this is just to study the electrical situation before you go crazy tracing leaks.

[Mark]

  I recently rewired my amp and found something funny that I can't explain. After I rewired it the gain pot sounded scratchy, not like a dirty pot but like a presence pot so I figured there was DC on the pot. I checked and there was ~.2V so I figured it was a leaky cap. I desoldered the end of the cap going to a 470K resistor on the board going to the gain knob and tested. There was no voltage on the cap but there was voltage on the pot. I disconnected the pot from the board giong to the 470K and there was no voltage on the pot. I checked the voltage at the 470K resistor which was now connected to nothing but two eyelets on the board and found ~.4V. I then desoldered the 470K resistor and there was ~.4V on the eyelet. I checked for flash or a loose piece of wire on both sides of the board to see if something was making a connection but there was nothing. I can't figure this one out. The eyelet that's giving me the .4V is about 3/16ths away from the 250V B+ for the preamp tubes. Is this too close? I can't figure out how there's voltage on that eyelet. Like I said there is no flash or drips of solder or strand of wire connecting the two on either side of the board. I don't think I'm getting a false read on my volt meter because the pot sounds scratchy like there's DC on it. I measured with an ohm meter that goes to 20M and there is no connection. The only thing I can think of is that the eyelets are too close together and the board is slightly conductive. Is there a minimum distance to how close you should have the eyelets. Any opinions, this is driving me crazy. By the way the fiberboard is from Mojo.

I bought a new fiber board from eBay that was conductive. I didn’t have a megger to test it. I used the voltage from my amp. I was measuring various voltages across the board. The stuff is unusable in my opinion.


Regards


Mark