Ceriatone Forum

In the middle of a gig, I suddenly noticed my clean wasn't clean anymore.
It wasn't really dirty,  just not as clean as it was before. 
Swapped power tubes and V1, to no avail.
When I got home, I put back the old (JJ) power tubes and measured the bias.
One tube looked normal at about 36mV, but the other tube measured around 70 !
I know the bias was fairly well balanced when I installed the tubes.
I tried the other pair (Mesa) of power tubes again, and found the same imbalance.
I ended up using 1 Mesa and 1 JJ and the bias for 1 is 35 and the other reads 41.
Both sets (JJ & Mesa) were bought as matched pairs, and as I said, when first installed biased correctly.

I built the amp, but I'm not a tech - just a monkey that can follow instructions.
Any ideas or advice ?

Are you able to drop the chassis and check the voltages on your tubes, specifically pin 5? Sounds like the bias resistors may be going out or maybe the screen resistors (pin 4)

Gregg

Thanks for the (quick) response.   Yes, I think I can manage that.  According to the chart I have, pin 5 should measure -42 V and pin 4 should measure 441 V, is that correct ?

I'll open her up and see what gives.  Seems strange that a supposed matched pair are so grossly unbalanced.  I just found 2 tubes that got a closer balanced reading - but I'm still getting some distortion on the clean channel - though its subtle - not really noticeable at bedroom level, but noticeable at anything higher.
THe OD channel is either unaffected, or the added distortion is too subtle to detect on the OD channel.

Pin 5 will vary depending on his much negative voltage is being applied to bias the tubes, but the should be the same on both tubes. Pin 4 should be about the same on both tubes.

Gregg

Is the voltage on pin 5 dependent on the tube that's in the socket ?
Right now the bias is reading about 41 in V4 and 35 in V5.
I would get different readings if I swapped those tube positions.

So if the bias is not equal, should pin 5 still be equal ? Or within what tolerance ?

Pin 5 should be the same irregardless of which tube is in which spot. The current draw has more to do with the screen and plate structure of the tube, along with the cathode, and can be effected by the screen resistors going out as well. In other words, if the rest of the tubes you are trying aren't way out of whack like the original set, everything may be fine, and the one drawing 70 mA may have been red plating and on its way out. How does it sound now with the better matched tubes?

Gregg

I don't think it sounds any better.  Both a matched pair of JJ's and a mached pair of Mesa's were way out of wack. I'm pretty sure the Mesa's have few hours on them.

FWIW, here's a couple of samples of how it sounded in the first set and the second set.

http://www.afgm.ca/JesusJustLeftChicago_Clean.mp3

http://www.afgm.ca/LittleWing_Distorted.mp3

I'll experiment tomorrow with the tubes I have and note the measurements.
It might be that I have 2 bad tubes in there. The one from each brand that was drawing a lot of current.
When I put the pair that's in there now, I had to turn the bias down quite a bit.

It seems no matter what, V4 gets noticeably hotter than V5.
I just tapped the tubes and got some noise, so I'm suspecting a cold solder joint somewhere.
Or it might have been because the tube is bad.
Got to get some zzzz's now.
Thanks for your help.

Jon

I'd check the screen resistors and make sure they aren't melty or smell burnt. If they don't then I'd check the bias voltage on each tube, preferably with two new matched tubes from a reseller that tests them at the prescribed voltage. If that checks out, check the plate and screen voltage on each tube (pin 3 and the pin 4). The plate and screen voltage should be fairly close on the OTS, if not, something has gone awry. I'm not knowledgable enough to know if you can hurt the OT without destroying it, but if you can, I wouldn't doubt that red plating a tube could cause that. Hopefully one of the more knowledgeable guys will chime in. I will say, be sure to keep one hand in your pocket while testing, and be sure to have a load on the amp, either your speaker cab or a dummy load, and all controls turned down while testing. Also, it wouldn't hurt to change your PI tube. The PI is part of the output half of your amp and is the unsung hardest working tube in your amp. Most times it's the PI we hear breaking down in the power section when we overdrive the power end of the amp, and it handles alot of voltage, so it's a good idea to change it out and re adjust the trimmer when you change your power tubes

Hope this helps

Gregg

Thinking on it some more, it sounds like either the Mesa in V4 went kerplooey during the set, taking out the screen resistor, or the screen resistor went out during the set taking out the tube. Either way, I'd be checking V4. I'd still check the negative bias voltage on both sockets, though, just incase.

Gregg

As I was preparing to do some tube swapping/testing I disconnected my D-lator from the loop.
Other times I've done this, the sound would fade in and out for a while, then stay on.
This time, instead of fading in/out, the tone went from thin/trebly to normal, then settled on normal.
When I plug a small patch cord through the loop it sounds normal.

So I have a weird problem with those jacks, its seems. The change from in to out or trebly to normal is not instant, it fades over a second or two.

First Tube Swap Test
M1 is the Mesa and J1 is the JJ.

In the first configuration they're roughly balanced.
V5                             V4
M1=35.4                 J1=41.6

after swapping - grossly unbalanced
J1=26.5                  M1=51.5

So V4 is always higher and whichever tube is in there gets a lot hotter than V5.
Does that provide any clues ?

Yeah, could be the screen resisitors, the bias resisitors, or both. I'd check the readings on the bias pins (#4) and see what you get. you are defintely pulling more current on V4 for some reason.

Gregg

Here's what I got:

Pin 4 optimum = 441
V5                      V4
P4 = 468            P4=467

Pin 5 optimum = -42
P5= -45.7             P5= -45.7

So voltages are all a bit high, but basically equal on the 2 tubes.

On pins 1 and 8 I get the same readings as the bias points.
The tube in V5 is barely warm while the tube in V4 is burning hot.

More info -
The bias resistors (1 ohm)

The one for V4 reads about 1.3 ohms.

The one for V5 seems dead - infinite ohms - no signal getting through
No visible sign of damage or burn on it.

This would indicate to me the problem is V5, not V4.
Tubes are supposed to get fairly hot,  no ? 
For a power tube to be barely warm so you can leave your hand on it forever doesn't seem right, now that I think about it.

But would that bias resistor affect the tube's performance and temperature ? Or just the bias measurement ?

Did some more research and now I'm sure the problem is with V5.
With any tube in V5 and no tube in V4, I get no sound.

Seeing as how the bias current gets to the tube through that 1 ohm resistor and its open, the current's not getting there, even though I get a reading.

So, I'm wondering - can I bypass that 1 ohm resistor and see if that fixes the problem ?
V5 should then get sound even if V4 is pulled out ?

What I don't understand is - if that resistor is creating an open circuit, where are the mV on pins 1 & 8 coming from ?
Why don't pins 1 & 8 read zero mV ?

The 1 ohm resistors provide a way to safely read the bias current with a DMM, and connect the cathode (pin 8) to ground. Instead of reading the actual current, your are reading the voltage drop across that resistor. Through the magic of electrical engineering and math that I barely grasp, the mV reading equals the current reading but is safer to measure. If V5's sensing resistor is bad it could cause a problem, as pin 8 and externally only pin 1 are the cathode and connected to ground. I would check the external connection between pin 8 and pin 1 along with changing that 1R 2W resistor, just to be safe.

Gregg

THanks Gregg,
I (sort of) understand what the 1 ohm resistor is for - converting mA to mV so you can easily measure the bias.
What I'm wondering is if I can bypass that resistor with a piece of straight wire to see if that gets V5 working again.
I know I can't measure that side in that case.

I want to see if the problem is just the resistor, or there's something else going on upstream from there.

Pins 8 & 1 are connected together and are getting voltage - the same reading as at the bias lug.  around 35 mV.
And it changes as I adjust the bias pot. I don't understand how that's possible if the resistor is not passing any current.

Or is it just that pins 1 & 8 are not grounded ? Could that be why V5 is not firing up ?

If it wasn't grounded, you would be getting zero current flow across. What kind of voltage are you getting at the plates (pin 3)? Also you said you unplug and plug your c-lator with the amp powered on? Is it on or off standby? Generally you shouldn't connect/disconnect that unless the amp is powered down or at least in standby. Also, when you do your testing are all your levels turned down?

Gregg

I'm getting 469 V on pin 3. Master Volume is turned off.

Unplugging the C-lator with the amp on is ok if the volume is low. Its never been a problem - I don't think connecting/disconnecting it can harm the amp.  Maybe blow a speaker if the volume is up.

What I need to know, if you can tell me, is if I can safely bypass that 1 ohm resistor just to see if that 'fixes' the problem.
If so, I can replace the resistor myself.
If its not safe to bypass, or if the bypass doesn't fix the problem, then I'll probably have to take it to a tech, which I can't really afford right now.

Yup, you can run the cathode straight to ground, just can't test the bias on that one without some different equipment, but you can bypass it.

Gregg

Muchos Gracias, Gregg. I'll give that a try.

Bypassed the 1 ohm resistor and all is well.  Going to get a replacement now.
Thanks Gregg.  You saved me a small bundle.
Cheers,
Jon

You're welcome. Sometimes it IS the simplest problem and fix! :)

Gregg

Hopefully, it won't happen again.  Its unusual for a resistor to just quit with no sign of burn, isn't it ?

Anyway, another question - really for Pickmaster, but since we're already talking and you may know the answer....

While at the electronics store I bought some extra 100K and 68K resistors to try Pickmaster's slope resistor mod.
I couldn't find any brown ones like the one that's in there, so I just got plain carbon fibre, 2 Watters.
I believe the slope resistor is 3W.  I don't know if the brown ones are carbon fibre or what.

Can I put that 2 Watt resistor in there ?

I *think* 2 watts is fine from what i recall being told on here before, and that's what i got when i was planning to change the resistor out, because mouser was out of non-bulk 3 watters a the time. I picked up metal film ones from mouser just to be on the safe side

http://www.mouser.com/ProductDetail/Vishay-BC-Components/PR02000206802JR500/?qs=sGAEpiMZZMtlubZbdhIBIDqBBcd6vlUvTCvt5QvopJA%3d
$0.10 USD each, no minimum

They have the 3 watters in stock right now as well

http://www.mouser.com/ProductDetail/Vishay-BC-Components/PR03000206802JAC00/?qs=sGAEpiMZZMtlubZbdhIBIHDatFdBRARu3kAymlOBwNM%3d
 $0.46 USD each, no minimum

I wonder if something was wonky from day 1.   After replacing the 1 ohm resistor and putting back the pair of JJ's I've been using for the last year, the bias measurement was a lot closer from tube to tube.

Anyway, I did the slope resistor mod - with a temporary internal switch from 100K to 68K.  Been playing with the 68K and loving it!  Still have to see how it sounds at gig level, but sounding rather stellar at home.
I'm also running the bias a bit cooler than before.  Found a bias calculator online - and now that I know the voltage on the pins, I can bias more accurately.  Apparently, the higher the voltage, the lower the bias setting to get the same dissipation.

The amp is starting to sound like what I bought it for.   Can't wait to gig with it this weekend.

Thanks again to  plasticvonaband, Pickmaster and others who've been so generous with their knowledge and skill.