Bad power tube, filter cap, resistor?

nickmm

Tele-Holic
Joined
Dec 22, 2010
Posts
899
Location
Austrailia
Well, I have the new matched pair but it seems like the one older one (older one is only like a couple weeks old) paired with one of the new ones gives a closer reading on the dissipation. Fortunately neither combo is red plating, that I can see.
If you swap the bad tube does it still read high dissipation? Or is the one slot always giving you the high reading?
 

joulupukki

Tele-Meister
Joined
Nov 26, 2020
Posts
348
Location
Utah
If you swap the bad tube does it still read high dissipation? Or is the one slot always giving you the high reading?
I am no longer using the bad tube. Not only does it make all sorts of bad noises, it red-plates regardless if it’s in V4 or V5.

I have a new tube in V4 and the old tube that was in V4 now in V5. I haven’t tried switching them around.

Next week just to see what happens I’ll try swapping the two new matched tubes in V4 & V5 just to see if I get the same values:

1656829581321.png
 

nickmm

Tele-Holic
Joined
Dec 22, 2010
Posts
899
Location
Austrailia
Tha
I am no longer using the bad tube. Not only does it make all sorts of bad noises, it red-plates regardless if it’s in V4 or V5.

I have a new tube in V4 and the old tube that was in V4 now in V5. I haven’t tried switching them around.

Next week just to see what happens I’ll try swapping the two new matched tubes in V4 & V5 just to see if I get the same values:

View attachment 1000410
Looks good to me play the amp now and see how it sounds.
Use the Old and New they are pretty close.
If you want to be on the safe side, up the bias resistor but compare the sound..
Tube life is a bit random these days.
 

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
8,454
Location
Salt Lake City
Good ideas here. FWIW perfect tube matching is not needed; some degree of mismatch increases nice harmonics.

Also @nickmm is saying more clearly what I was hinting at, your bias may just be too hot. It's been a busy trip, and I forget, but were your prior tubes also JJs, matched, and burned in? Did the bad one head south when you cranked the volume? Why will these be different? How bad would it sound biased at 100%? Etc. :)

OTOH I’m also confused by the 'new math' suggesting your MPD is much lower than before. I’ll let our smart friend @andrewRneumann work on the math…
 

joulupukki

Tele-Meister
Joined
Nov 26, 2020
Posts
348
Location
Utah
Also @nickmm is saying more clearly what I was hinting at, your bias may just be too hot. It's been a busy trip, and I forget, but were your prior tubes also JJs, matched, and burned in? Did the bad one head south when you cranked the volume? Why will these be different? How bad would it sound biased at 100%? Etc. :)
Yeah, I’ll have a 150 Ohm 5W resistor next week at some point and will install that instead of the 130R one. I’m guessing that’ll reduce the dissipation a bit and help extend the tube life.

The first set of JJ tubes that came with the Mojotone kit were (according to them) burned in and matched. It played well for a couple weeks and then when everyone was out of the house I hooked up my wireless transmitter to my strat, set it to the bridge humbucket, and with ear muffs on cranked it all the way up and played it loud. Good times were had until at some point it dropped volume and started acting odd. At first it would only make a whooshing and popping noise for a few seconds at startup and then seem fine but after a couple days of that the whooshing/popping wouldn’t stop and that’s when I discovered that one of them was red-plating … and causing the whole thing to have bad voltage & current measurements. Before that incident it was all working pretty well.

OTOH I’m also confused by the 'new math' suggesting your MPD is much lower than before. I’ll let our smart friend @andrewRneumann work on the math…
I’m guessing that the reason why it’s different is that from the first measurements is because it’s using a different way of measuring it all. All of my first measurements were done using the steps shown in Rob’s “Measure Bias With the Output Transformer Resistance Method” which depend on measuring the resistance of each output transformer secondary lead to OT center tap soon after shutting the amp off (with it still warm). Those resistance values seem to change depending on when I measure it.

With this other way of measuring, the resistance of the OT secondaries isn’t used in any of the calculations. Instead it’s using the 10R ohm resistor on V4, measuring the voltage across that resistor, dividing by 10 to get the current, measuring the voltage across the shared cathode resistor, subtracting V4’s current from the total to get V5’s, etc. It would be nice if the two different ways matched up more nicely, definitely. But, I am guessing this method is probably more accurate?
 

nickmm

Tele-Holic
Joined
Dec 22, 2010
Posts
899
Location
Austrailia
Just use the tested methods. Ken Fischer knew something.
Don't exceed 100% dissipation that is what the current manufacturers use, not sure NOS are the same tolerance.
sorry about the time difference.
 

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
8,454
Location
Salt Lake City
Glad you're upping the cathode resistor, and I'm sure you'll check bias before you crank the amp.

I'll wait for @andrewRneumann to respond about the 10R resistor -- seems like if it's to be useful it should *not* measure dissipation differently than the two Rob-calculator methods, OT resistance and cathode resistor voltage drop. But I don't understand the wiring or the math on this new method, so I'm no help.
 

andrewRneumann

Friend of Leo's
Joined
Mar 22, 2020
Posts
2,031
Location
Cincinnati, OH, USA
Glad you're upping the cathode resistor, and I'm sure you'll check bias before you crank the amp.

I'll wait for @andrewRneumann to respond about the 10R resistor -- seems like if it's to be useful it should *not* measure dissipation differently than the two Rob-calculator methods, OT resistance and cathode resistor voltage drop. But I don't understand the wiring or the math on this new method, so I'm no help.

Anywhere we find a resistor, we can easily find current. We simply measure the voltage across the resistor and use Ohm's law to find the current. The problem we have before us is laid out below. We want to find Ia1 and Ia2. There is no resistor on Ia1 or Ia2, so we have to figure it out indirectly.

We do know the current at all the resistors. So Is1, Is2, Ia1 + Is1 and Ia1 + Is1 + Ia2 + Is2 are all known quantities. It is algebra to use those four known quantities to determine Ia1 and Ia2 by themselves.

Ia1 is easy. Measure the voltage across R1 and R3. Convert them to currents using Ohm's law. Then subtract Is1 from Ia1 + Is1 to arrive at Ia1.

Finding Ia2 requires one more step. Measure and use Ohm's law to find Ia1 + Is1 + Ia2 + Is2 and subtract out Ia1 + Is1 and Is2 to arrive at Ia2.

The resistor R1 should be 1Ω to keep from causing too big of an imbalance in V1 and V2. It can act as the jumper between the cathodes of V1 and V2.

1656879345275.png
 

joulupukki

Tele-Meister
Joined
Nov 26, 2020
Posts
348
Location
Utah
Glad you're upping the cathode resistor, and I'm sure you'll check bias before you crank the amp.
For sure. It’s not gonna get cranked ‘til I know it’s more safe. No since ruining another tube.
I'll wait for @andrewRneumann to respond about the 10R resistor -- seems like if it's to be useful it should *not* measure dissipation differently than the two Rob-calculator methods, OT resistance and cathode resistor voltage drop. But I don't understand the wiring or the math on this new method, so I'm no help.
The thing that makes the OT resistance method seem inconsistent to me is that the resistance changes depending on how warm the amp is. I mean, I took multiple measurements across the last few weeks of the OT resistance and I ended up with different values over and over (measured from Plate to OT Center Tap, v4 & v5, respectively): (327, 340.3), (322.5, 333.6), (322.5, 339.7), (336, 319).

For now, I removed the 10R resistor to keep things clean. I had disconnected the cathode wire from V4 to add the resistor as a jumper in temporarily. I did not know about the 1R trick/tip until after I had made an order of stuff … too bad there’s not an electronics store ‘round here.
The resistor R1 should be 1Ω to keep from causing too big of an imbalance in V1 and V2. It can act as the jumper between the cathodes of V1 and V2.
Nice visual with great explanation. Thank you! When you use the 1Ω resistor as a jumper are you flying it over the top of the tube sockets? I should have taken a picture of mine while I had it in place.
 

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
8,454
Location
Salt Lake City
It is algebra to use those four known quantities to determine Ia1 and Ia2 by themselves.

Wow, mad math skillz, Andrew, but even better -- awesome visuals. Thank you, really. You put some work into that.

Back in King Fan Kindergarten, I said above I'd never thought of using 1Ω resistors in cathode bias though I do it in fixed bias all the time. I guess the only minor issue in just hanging one on each cathode is whether you have an empty pin to span them to? If you're going to be probing across 'em, seems like hanging in space isn't perfect.

Still. It may be obvious why small, added, individual 1Ω 1% cathode resistors are better than using the big ol' 250-470 cathode resistor you've already got, but not to me. Better cuz they're individual? you don't have to average out the resistance seen by the two cathodes? I figure it can't be *that* much better just due to the 1% precision, since you can measure the cathode resistor.

took multiple measurements across the last few weeks of the OT resistance and I ended up with different values over and over (measured from Plate to OT Center Tap, v4 & v5, respectively): (327, 340.3), (322.5, 333.6), (322.5, 339.7), (336, 319).

Those are resistances, amp off, in ohms? Or amp-on voltages as the amp warms up? Apologies, since I don't use this method, but Rob's 5E3 OT has resistances in the range of 70-80Ω. Rob does say a warmed-up (then turned off) OT will give the most accurate values, but notes it shouldn't vary -- it's a fixed physical resistance in lengths of wire.
 
Last edited:

andrewRneumann

Friend of Leo's
Joined
Mar 22, 2020
Posts
2,031
Location
Cincinnati, OH, USA
Nice visual with great explanation. Thank you! When you use the 1Ω resistor as a jumper are you flying it over the top of the tube sockets? I should have taken a picture of mine while I had it in place.

Yeah. If I was worried it might short something I would would heat shrink the leads. Remember that all these methods of measuring resistance/ current are only as accurate as the meter we are using. Resistance in leads / probe leads can make a difference too. Most of the time we don’t need that granular level of precision. Quoting dissipation out to 5 significant digits gives a false sense of accuracy.
 

andrewRneumann

Friend of Leo's
Joined
Mar 22, 2020
Posts
2,031
Location
Cincinnati, OH, USA
you don't have to average out the resistance seen by the two cathodes?

This is the only reason to do it with the extra resistor(s). Some people just want to know what each tube is drawing. Averaging isn’t good enough. :cool: Got to be above average. I don’t really advocate doing this, but the thread was going in this direction and the OP was talking about mounting an additional terminal strip which I didn’t feel was necessary.

If one were to incorporate a fixed/cathode bias hybrid, one could actually do something about the imbalance besides buy a better matched set of tubes. I believe other smart people have posted ideas along this vein.
 

joulupukki

Tele-Meister
Joined
Nov 26, 2020
Posts
348
Location
Utah
Those are resistances, amp off, in ohms? Or amp-on voltages as the amp warms up? Apologies, since I don't use this method, but Rob's 5E3 OT has resistances in the range of 70-80Ω. Rob does say a warmed-up (then turned off) OT will give the most accurate values, but notes it shouldn't vary -- it's a fixed physical resistance in lengths of wire.
Correct, resistance measured in ohms with the amp off after having been on for a while. Measured V4 and V5 from B+ where the OT primary connects to V4 and V5 Pin 7, the OT secondary. I also saw that he said you should only have to measure it once. But, as I've been testing I've measured it the different times and it seems different each time. Dunno why. Probably the heat affects it?
This is the only reason to do it with the extra resistor(s). Some people just want to know what each tube is drawing. Averaging isn’t good enough. :cool: Got to be above average. I don’t really advocate doing this, but the thread was going in this direction and the OP was talking about mounting an additional terminal strip which I didn’t feel was necessary.
Hehe. Sorry, must be the engineer in me. Just really surprised how much emphasis I see all over about how important matched tubes are only to find out the plate currents aren't really the same in practice anyway. Odd.
If one were to incorporate a fixed/cathode bias hybrid, one could actually do something about the imbalance besides buy a better matched set of tubes. I believe other smart people have posted ideas along this vein.
Yeah, I think I'll just add a little higher of a shared cathode resistor when I get it, call it good, and enjoy playing this beast.

Thanks for the exploring this idea with me. :)
 

2L man

Tele-Afflicted
Joined
Nov 23, 2020
Posts
1,549
Age
62
Location
Finland
I check that tubes are somewhat balanced and then measure hotter tube safe max bias and mark this trimmer position. Then I play and adjust the bias. I have also used an oscilloscope but sometimes I have set bias bit lower accepting more crossover distortion. Good sounding bias might be different for loud and quiet use amp?

So bias can be seen as an adjustment like other potentiometers!

Usually tubes bias drift, some a lot but some not, which is good to understand that setting once and forgot does not always work.

Defining max bias is vacue but for higher OT load=lower impedance and higher operative voltages/stiff power supply where loadline pass the max plate dissipation very soon it is good to limit max bias to prevent tubes getting too hot.
 
Last edited:

joulupukki

Tele-Meister
Joined
Nov 26, 2020
Posts
348
Location
Utah
New report…

Today I installed a 150R 5W resistor as the cathode resistor to bring down the plate currents a little bit. I’ve also got two different sets of matched tubes to tinker with: a pair of JJs and a pair of Sovteks.

This time I also let the amp warm up for a good 30 minutes and *THEN* took the resistance measurements on the OT primary leads. I also compared it with the shunted OT mA measurement and the plate current readings are pretty close – enough to where I’m convinced the OT resistance method is pretty accurate. During one test I calculated 36.7 mA with the OT resistance method on V4 and the actual plate current measuring it with the leads between the plate and OT center tap was 36.02 mA.

OT Resistance (V4, V5): 340.2, 322.9
5W Cathode Resistor: 148.4

Measurements for the matched Sovteks (V4, V5):
Cathode Current: 10.69 V
Plate Current: 32.0 mA, 32.5 mA
Plate to Cathode Voltage: 320.6 V, 321.6 V
Plate Dissipation: 10.27 W, 10.46 W
Plate Dissipation %: 85.6%, 87.1%
Rob’s Tube Calc via Cathode Resistor Voltage Drop: 90.8%

Measurements for the matched JJs (V4, V5):
Cathode Current: 10.67 V
Plate Current: 32.6 mA, 32.5 mA
Plate to Cathode Voltage: 320.6 V, 321.5 V
Plate Dissipation: 10.46 W, 10.45 W
Plate Dissipation %: 87.2%, 87.1%
Rob’s Tube Calc via Cathode Resistor Voltage Drop: 90.8%

For now, I’m gonna run these new JJs and see how it goes. It is still plenty loud, sounds great, and puts out some gnarly good distortion. Since both of these new sets of tubes match so closely in the readings it makes me wonder if that other JJ set wasn’t matched like they said it was. Next time I can make it in to the local tech who has a test machine I’m gonna have them double-check them just for curiosity’s sake.
 

Lowerleftcoast

Poster Extraordinaire
Joined
Dec 30, 2019
Posts
5,646
Location
california
The thing that makes the OT resistance method seem inconsistent to me is that the resistance changes depending on how warm the amp is.
Iirc, Rob mentions to measure the resistance when the OT is at operating temp. If he didn't he should have. Once the OT resistance has been measured at operating temperature, that measurement can be written down and used in the future. It should not change if the operating temp is the same. If the OT is damaged in the future, this resistance may be used to identify that fault.

To determine the accuracy of the tube match, simply swap the tubes and re-check the numbers. If there is no change, the tubes are very well matched. There are tolerances for tube matching and as tubes age the current draw will change... don't expect perfect matches now or a month from now.
 




Top