|Mr_bibbles||6/1/2018 2:42 PM|
|Cathode Bias Red Plating Discussion|
Was working on a Krank Rev Jr. recently, tubes are red plating. I would post a schematic but I can't find one anywhere. It is a cathode biased amp, seems pretty simple in terms of the OT tube stage, Cathode biased 6L6, can be subbed with 6v6 according to Krank.
Increased the Cathode resistor from 500 Ohms to 1k (quick and dirty test...) no change there. Tried with (several) known working tubes, they all red plated. (These were crummy test tubes FYI).
The Plate / Grid voltages are:
Pin 5 Control - 30mv
Pin 4 - Screen - 267VDC
Pin 3 Plate -243VDC
Pin 5 Control 0mv
pin 4 Screen 411 VDC
pin 3 Plate 411
From my limited knowledge, I am under the impression that Screen voltage is kept slightly below plate voltage to help control the current in the output tubes. if this is true, could that be the source of redplating for the tubes?
And further, can someone help explain the relationship between the screen and it's control circuit? Without a schematic (and with an untraceable, unremovable PCB) I'm at a bit of a loss.
Thanks in advance,
|The Dude||6/1/2018 3:48 PM|
|You didn't report the cathode voltage which, IMO, is the important thing. Measure voltage across the cathode resistor and calculate bias. Is there a cathode bypass cap? Maybe it's shorted. Measure the in circuit resistance across the cathode resistor. If you measure less than your cathode resistor, my bet would be a shorted cathode cap. Of course all conjecture without a schematic.|
|Mr_bibbles||6/1/2018 4:51 PM|
Good guess! Measured 345 ohms across the resistor in circuit, measured 500 ohms with one leg up. There is no bypass cap connected to the resistor or near it, I did however bridge the resistor with a 100uf Cap in parallel to see if that would help, and it did not. The traces are on the underside and the sockets prevent me from removing the board without significant work. Bonus aggravation is polyester block caps/radial electrolytics who's leads I can't measure for continuity or anything else. I also already contacted Krank and they cannot get me a schematic...
Cathode resistor voltage drop (Measuring at both sides of resistor with tubes in) 1.985VDC, which according to Weber Bias Calc puts the tubes at 5% of 6v6 dissipation at 290 Plate to ground voltage??? This is either crazy low or I am doing it wrong. Without access to a schematic, would there be a way to hook up a Bypass cap to correct the short? Or is there some way I could make an educated guess on the electrolytics around the OT pins??
Cathode resistor measures low in circuit, fine when lifted. Cathode bias seems low but I'm not 100% sure I calculated it correctly. Bypass cap possibly shorted, difficult to find out which one it is.
|The Dude||6/1/2018 5:05 PM|
|Adding another cap won't remove the short (or low resistance). You'll have to hunt it down. Try removing the output tubes and checking the resistance again. You could have a tube with an H/K short. If that doesn't change things, you'll have to go deep diving. When you calculated bias current, did you use the measured resistor value or the value stamped on the resistor? The cathode cap, if there is one, would likely be in the 25-50V range. It wouldn't be one of the larger 350-500V caps. Since you don't have a schematic, can you upload a picture?|
|Mr_bibbles||6/1/2018 5:36 PM|
|Makes sense for the cap not making a difference, duh! 0 ohms (short) parallelled with anything will be 0.|
Measured, meaning in circuit value? 258 Ohms, plugged it in with same voltage drop and P --> C voltage, 9% per tube?
I've tried multiple tubes, so I feel like I can rule out the 6v6's. What is an H / K short though?
here are some pics:
There are really only 3 appropriately sized polarized electrolytic caps for the bypass, I could just snip / top mount them all.... guessing it's one of the two on the right though. The one by the FX loop (left side) is 100uf, the other two are 22uf.
|eschertron||6/1/2018 6:17 PM|
|my hypothesis: something (I can't say what) is shorting out the cathode resistor, or seriously reducing its ability to control tube current. You've tried replacing the cathode resistor? Good. Now unsolder it and remove it from circuit. Do the tubes still redplate or show the B+ being pulled down?|
edit: could a lead from the cathode resistor be touching the chassis?
|Chuck H||6/1/2018 6:46 PM|
|Well you said the cathode resistor measured 345 ohms in circuit and then you said it measured 258 ohms in circuit. If both of these are accurate then something other than the cathode resistor is kludging the resistance. I like escher's idea of lifting the resistor and rechecking. That would tell you if you need to hunt down a short real fast.|
|Enzo||6/1/2018 9:15 PM|
|Pull the tube, power up, any voltage on cathode pin?|
Lift 500 ohm resistor, now measure resistance to ground from cathode.
Is cathode pin touching heater pin?
|Mr_bibbles||6/2/2018 12:06 PM|
I poked around the circuit and found another 500 Ohm 5w Cement resistor near the filter caps, My guess is that the designer (or some previous tech) had used 2x 500 Ohm resistors in parallel, creating a combined 250 Ohm cathode resistor that would jump to 500 if one fails?
Resistance from cathode to ground becomes 500 when unsoldered, see above about two parallel resistors?
Cathode shows OL to all pins except 1 ---> 8
So, based on this information, we can rule out a short from Cathode to Ground? Leaky Coupling caps from the previous stage maybe?
As always, thanks boys.
|Enzo||6/2/2018 4:54 PM|
|Nothing to rule on, you found a parallel resistor.|
An amp designer would not use two different 500 ohm resistors in different places around the amp but wired together.
|Chuck H||6/2/2018 7:13 PM|
So... If 250R is a typical cathode resistor in similar conditions, why the red plating? And with the two 500R resistors located remote from each other, just what's really going on with the design? A schematic would really help and may be necessary.
|Enzo||6/2/2018 8:11 PM|
|They were not placed together, but he said they were wired in parallel, that kind of together. Wired together in their separated placces.|
|Chuck H||6/2/2018 8:37 PM|
Still... 243Vp and 267Vg2? That's a wider than typical discrepancy and with an unloaded B+ of 411V along with red plating and the the cathode resistance (pretty much) checking out I think it's reasonable to suspect the plate loading. It could be a board short, the OT, a bad speaker cable, speaker, jack, etc. But I think something is loading the plates into too low an impedance rather than this being a bias issue.
|dstrat||6/2/2018 9:17 PM|
|assuming this is a single tube amp... its no wonder tube is glowing, bias too hot!! should have 500 ohms R |
suggest remove one of those 500 ohm resistors... 250 ohms works for Two Tubes!
the cathode voltage on pin8 would answer the question.. methinks
|Chuck H||6/2/2018 10:07 PM|
|g1||6/3/2018 12:11 PM|
|Can you lift both those cathode resistors and check resistance from pin 8 to ground?|
Also power up and see that it is not red-plating (with both resistors lifted) ?
If you still have a problem, then I think there is a board issue. If no more problem, probably plate loading as Chuck mentioned in post #13.
|Mr_bibbles||6/5/2018 11:53 AM|
|Thanks everyone, to clarify there are 2 x 6v6 in the power section. I think another tech was working on this before me, it's possibly that he added the second resistor- it's wired straight from the pin to ground instead of to an appropriate spot on the PCB. Not sure on this though.|
It seems like the two resistors are in fact wired in parallel, when I removed both of them from the circuit and powered it on I still get red plating. Resistance from cathode pin 1 and 8 to ground with both removed became OL. (I measured it a few times, sometimes it showed OL, sometimes it shows 7Mohms, sometimes it shows 2MOhms..)
So in summation-
There are two Resistors wired in parallel, Possibly a modification, possibly intentional. There is no short from cathode to ground, but the tubes are still red-plating even with both resistors removed from the circuit.
Possible next steps- Something is pulling B+? Can we rule out a bypass capacitor short with these measurements?
|Enzo||6/5/2018 1:22 PM|
|Red plating tubes means there is a current path through them. If you measure open to ground at the cathodes (2M or 7M or any other really high reading doesn't matter - that is open for us), then look to see if a tube is shorted internally. ANything shorting to heater?|
|eschertron||6/5/2018 2:12 PM|
|How about a low resistance to ground from grid 2? Seems unlikely, but with 'surprise' resistors, anything else is possible. Is there something special about the screen circuit in this amp? The red-plating is on both tubes, correct?|
edit: I guess this is really a question: can grounded screens cause the plates to glow? Or do the screens melt first?
|eschertron||6/5/2018 2:17 PM|
Also, have you checked with different tubes, after removing both resistors and verifying no continuity to ground from the cathode pin?
|Enzo||6/5/2018 2:18 PM|
|I think grounding the screens is a good way to shut off current in the tube.|
|eschertron||6/5/2018 2:30 PM|
|Which is why I asked, not sure if the cathodes floating makes a difference. Now there's not a ground-potential 'target' the size of the side of a barn, and the screens could be exposed. I'd expect screen dissipation could work both ways|
|g1||6/5/2018 9:32 PM|
|I'm wondering if there may be a pc board leakage resulting in cathode short to ground only when powered up.|
If I read correctly, the voltage measured at the cathode was always close to 0 VDC.
|Chuck H||6/6/2018 7:04 AM|
|B+ is dragged down hard Red plating with an OL on the cathode When cathode resistor is in there is less than 2V standing Amp does this with several subbed tubes This one is a corker. But I'me going out on a limb and saying the answer is in the measurements, and likely how those measurements were taken. I'll guess that your cathode bypass cap is shorted, but only when the amp is on. Your meters test conditions may not be sufficient to induce the short, which may be internal arcing. The very next thing I would try is lifting the cathode bypass cap.|
|Chuck H||6/6/2018 7:34 AM|
|I'm thinking the cathode bypass cap must be shorting with operating conditions and the DMM test parameters don't induce the short. The next thing I'd try would be to lift the cathode bypass cap OR just replace it if it's a PITA taking the board out and putting things back together. A new cap won't hurt if that saves labor.|
|Mr_bibbles||6/7/2018 4:09 PM|
I repeated the experiment with another set of crappy,known working from another amp, test tubes (I have a whole box of 6l6's...) They both redplated, but one of them started much faster (I'll call this tube A). When I swapped sockets, they both redplated again, and tube A started first again in the other socket. Guessing that just means they're off by a few MA..
|eschertron||6/7/2018 5:07 PM|
|Have you taken pin voltages with tubes in, under power, since you removed all components in the cathode network?|
Did you say you were going to remove the cathode bypass cap? Is that out as of the last reported test?
...I think I mis-spoke when I asserted the cathode pin was not the source of the ground potential. Forgive me if I got ahead of you.
|Chuck H||6/7/2018 8:15 PM|
|Cathode bypass cap.|
|Mr_bibbles||6/8/2018 4:26 PM|
There are two diodes in the circuit, d3 and d4. I reconnected the cathode resistors, lifted the leg of one of the diodes, taking a guess that it might be a flyback / snubber / whatever you call it. Red plating has stopped, sound is good, a little buzzy. cathode of the Diode is tied to ground, anode tied to both pins 1 and 8. It's a IN4007, so I replaced it with a known good one. The red plating has come back!
So what is the function of that diode, and how could it be tied to the bias? Could it be possible that there is an underboard short to this diode, forcing current through ground from the cathode, effectively shorting the cathode ---> ground
|g1||6/8/2018 5:48 PM|
|What is the other diode, and what is it connected to? I was wondering about maybe a zener somewhere and if it could be shorted.|
|Mr_bibbles||6/8/2018 6:45 PM|
|I thought it might be another flyback, but it's not connected to cathode. Also, I don't even know if you can use a flyback for cathode bias.|
The anode goes to x2 paralleled 1k resistors, which goes to the grid. The cathode is tied to the legs of two resistors, one 24k and one 10k, but I cannot find where the path goes after the resistors. Both are close to what I assume are the Coupling capacitors for the PI.... part of a long tail inverter???
Note- this Diode DOES measure well in/out of circuit. Numbers are a bit scratched from removing it, but looks like 400-X- . In4007?
Any thoughts on the other IN4007 tied to cathode? Considering that removing it stopped the red plating but otherwise sounded passable at the very least, that seems to be a prime suspect?
|Chuck H||6/9/2018 5:59 AM|
|I'm afraid a schematic is in order as it seems there is an unusual circuit here. Trying to analyze it based on typical circuits (ie: Rk, Ck) is useless now and, I'm sorry, but the diode implementation is unclear.|
A diode with it's cathode to ground parallel to the cathode resistor pretty much moots the cathode resistance and the power tube cathodes are grounded. Without a negative voltage on the grids there is almost no bias. Ergo, red plating. The amp wouldn't have been designed like this.
Maybe it would help to know some back story on this amp.?. Do you know if it ever worked? Do you know more certainly that another person has been modifying it or trying to "fix" it? It's a contemporary design, was it beyond warranty when it failed initially? Did someone void the warranty by poking around in it or is it possibly stolen ?
Working from inside a box (knowledge of typical designs) trying to diagnose outside the box while waiting for Easter eggs of circuit info is becoming frustrating.
|Mr_bibbles||6/9/2018 12:19 PM|
Called Tony from Krank, he was not able to help besides telling me "it's cathode biased", and no schematics exist of this online. Another caveat is that the board cannot be removed unless you unsolder the tube sockets from the board- they were soldered in after the board was set in place. I think unsoldering them is my next move if we cannot determine what this diode is. Everything looks original from the outset, but it's possible another tech lifter / added a parallel cathode resistor. because it's wired directly from pin 1 to a connector pin tied to ground.
Chuck, I think I understand what you mean about the diode- if it's parallel to the resistor, current would flow with no resistance to ground, so no tech in their right mind would ever use one here. Perhaps there is a failed cap between cathode and this diode?? Or, as G1 astutely guessed a while ago, a board short.
Assuming it's SUPPOSED to be there, would anyone else have ideas about what the purpose of the diode to ground is?
|Chuck H||6/9/2018 12:55 PM|
|Here's a question, that diode that seems to be across the cathode resistor has it's cathode terminated at ground, so... I'm sure you tested that termination as ground, but is it supposed to be? Is it possible that diodes cathode terminates on a trace that shouldn't be ground? You can probably determine this by looking at other circuits connected to that trace. If it seems the trace shouldn't be grounded then you can look for the board short that g1 suggested or maybe another shorted component.|
|g1||6/9/2018 6:57 PM|
|Agree with Chuck and was also going to suggest you verify the diode connections by following the traces if you have not done that.|
If you were just going by resistance readings, it's possible there is something shorted between tube cathode and diode anode. Or something shorted between diode cathode and ground. Trace it out visually.
Also, does it look stock or like someone may have added it in?
|Mr_bibbles||6/13/2018 4:49 PM|
I think this is getting to the heart of the problem though, I'll be unsoldering the tube sockets and tearing it out of it's berth.... I'll let you know what I find in case anyone else has these issues.
Also, this diode is marked correctly on the board in print, factory installed.
|eschertron||6/13/2018 9:03 PM|
|I've had iffy (read: better than nothing) results tracing a board by sticking a high-intensity diode on a wand under the board. It may reveal just enough shadowy information to suss out some of the unknown connections. If the board's not fiberglass/semi-opaque then this won't work so well.|
|g1||6/13/2018 9:24 PM|
|Enzo||6/13/2018 9:27 PM|
|I had the same reaction... Some sort of sensing diode??? Then, oh... a light. I use a 100 watt bulb. I have a clamp on work light.|
|eschertron||6/14/2018 6:41 AM|
|Yes, typing past my bedtime leads to errors of omission. I'm thinking with the PCB still installed in the chassis, there might not be room to get a work light under the board.|
|J M Fahey||6/14/2018 9:52 AM|
|I was chiming in to say something and I see you are near, but not exactly there.|
I think that "somebody" who must be an avid "Forum reader" or a Graduate at "You Tube University" has added the "Ruby Mod" or whatever they call that dreaded invention of biasing tubes with a Zener diode.
FWIW it might even come from Factory itself.
In any case, either Factory or user added Zener diode shorted, Service "Tech" thought "oh! ... a diode!!! I have a few of those!!!" and replaced it with 1 x 1N4007 ... or 2 in series, same thing ... and oriented so they can pass current ("why would the original one(s) be oriented the wrong way"?) [headscratch] ... the perfect Storm.
Diodes get forward biased under operation , drop only 1 or 2V, and give power tubes all the current they want, so they happily redplate *and* overload +B down so much that it about halves and in a short time will also burn the PT.
Of course Multimeter, set to a resistance reading scale, does NOT show any diode(s) there.
Out of curiosity, I would measure cathode to ground, on the diode scale.
An alternative theory (hey!! I saw it at Alex Jones Channel so it must be true ) could be that you have there a properly oriented (replacement) Zener, but way lower Zener voltage, so tubes are always underbiased (voltage wise) and over biased (current wise). Pick one.
I wonder why the mains fuse does not blow, although I would not be surprised at finding it "reinforced".
So I suggest first of all remove those damn cathode diodes and plase sensible value cathode resistors.
Amp "should" work as normal.
|J M Fahey||6/14/2018 10:07 AM|
|Mr_bibbles||6/14/2018 5:07 PM|
But I'm having a terrible day. One of those days where everything you touch turns to shit.
When I (re-)removed the diode and powered it up I magic smoked the 1k (grid?) resistors in parallel. I have no idea how, It could have been the tubes, left a snipped lead in, left it diconnected for testing and forgot to re-connect... I repaired the damage (and managed to shock the shit out of myself in the process... like I said, terrible day...) and fired it up on the limiter, worked fine, added different tubes, worked fine, not pulling too much current. BUT now there is no signal / hum / buzz. Traced a signal through to V1, noticed there is NO PLATE VOLTAGE on any preamp tubes, but there is 350 on the plates of the power section. Open dropping resistor? any other suggestions? I noticed there is only 2 diodes near the primaries for a rectifier... they measure OL / 0.000 on the diode tester, but if this IS the rectifier and it WAS shorted, wouldn't the DC be missing from the output section?
On the plus side, the red plating is gone.
|g1||6/14/2018 6:13 PM|
|This should be unrelated. They would not have a separate rectifier for the preamp supply. Like you say, the power tubes are still getting plate voltage, so there must be a dropper resistor or fuse open somewhere.|
|J M Fahey||6/15/2018 3:18 AM|
Only you didnīt notice because bthe redplating tubes problem is way more dangerous and you (and probably an earlier Tech) focused on it only.
Hey, if at the same time a dog is biting your ankle and a mosquito is biting your ear ... what would be the focus of your attention?
I wouldnīt even be surprised that an earlier Tech disconnected the preamp trying to find what was pulling his +V down
|Zozobra||6/15/2018 3:48 AM|
|I had these lurking in my folder of random schematics. Hope they help elucidate the problem.|
|eschertron||6/15/2018 6:26 AM|
|I don't see D3 and D4 on the drawing. At least you have confirmed the two x 500R cathode resistors are stock|
|Mr_bibbles||6/15/2018 11:27 AM|
|Mr_bibbles||6/15/2018 11:31 AM|
|Zozobra||6/21/2018 4:34 AM|
|No problem. Hope they're useful.|
The krank's are interesting designs to dissect, at least in how they copied something and ended up with a semi-successful product. At their very core they are basically tweaks of the Sovtek Mig-100H with a dual recto-ish poweramp with no NFB and vox presence. This makes a lot of sense as the guy who ran krank used to import them and flip them on with a few modifications. The Mig circuit isn't one that has been copied a great deal as most people probably think they're cheap Russian crap but I rather like them. They aren't terribly subtle though!
The Mig-100H in itself is pretty interesting as all the Mig-100's apparently use the same PCB and they really kludged a conversion from a 3+CF topology of the Mig-100 to the 4 stage plate driven stack of the 100H. Notably the easiest way to achieve this was by referencing the 3rd to 4th stage divider to the cathode of the 3rd stage rather than ground. This has sparked a fair few arguments about feedback and bias shifting and what tonal effects this may have. I was certainly guilty of this until someone pointed out it was just kludge to get it to work on an existing PCB and that the designer probably didn't give a shit about either of those things; they just wanted a 4 stage high gain ripper! The kranks reference to the 4th stage cathode.
Some people find some parts of the krank design a bit questionable as the shift/sweep and presence controls do have HV on them. Both could have been avoided by either changing their placement or adding a cap without any real tonal penalty.