1961 B-15N, new problem with my power tubes and sparking...

#61
Thinking I'll just clean it up as much as I can with some alcohol and compressed air. I see that too... wasn't me! I haven't done much at all to the B-18, as you can see.

By the way, as you can see in the photo, even though a modern 3-prong has been installed, the deathcap is still there and connected. Any reason I can't just remove that lead on the left side and leave it in place? Disconnecting either side alleviates any danger, right? I'll be putting the new cap can in anyway, will just leave that wire off.
 
#63
agedhorse said:
I see chemical residue all over the backs of those power tube sockets... looks like somebody sprayed some "cure-all miracle cleaner" on them at some point.
Click to expand...

Yeh, and spraying on the BACK of tube sockets doesn't really help any regardless, does it? Maybe if it drips
all the way through :) ( not good )
 
#64
Personally, I don't like to spray anything on tube sockets. Most problems with sockets are related to the contacts tension weakening from heat, rather than oxide. plugging a tube in and out a few times should effectively scrape any oxide off the pins that might be there.

If you really still want to spray some contact cleaner in the sockets, ensure it's a zero residue cleaner.
 
#65
agedhorse said:
Without the tubes installed, you better have zero volts on the grid. If you don't, I would immediately suspect the .02uF blocking caps. If one is shorted, I sure wouldn't want to discover this with a set of tubes present... could be responsible for the op's symptoms (if that's what's measured).
Click to expand...

Agreed - this was the primary problem I had originally with mine. I replaced them both.

I redid the PS caps that were a tad leaky and then experienced a flash over in the 5AR4/GZ34 rectifier. I followed a tip from @beans-on-toast and installed silicon diodes in series with the 5AR4 plates. That solved that. Then an output tube red plated so I replaced it. The other then did the same so now it’s running happily with two old tubes from my stash.
 
  • Like
Reactions: beans-on-toast
#67
Frederick34 said:
Any reason I can't just remove that lead on the left side and leave it in place? Disconnecting either side alleviates any danger, right? I'll be putting the new cap can in anyway, will just leave that wire off.
Click to expand...

You can just remove it, I like to replace with a safety cap so you can have the line filter if you ever need it. There's a pretty thorough writeup in the portaflex wiki by BoT but this is the cap you'd want to use.
 
#68
That cap is not a line filter, it simply attempts to reference the chassis to one side or the other of the AC line via capacitive reactance. The idea is that one side or the other may be bonded to earth grind, and that may help to reduce noise.
 
  • Like
Reactions: coreyfyfe
#69
Rather than call it a death cap, I like to call it an AC ground capacitor, years ago they were called line filter capacitors. The capacitor does a couple of things, it provides an AC ground reference which can shunt RFI in the power line to ground, it also limits the AC current on the chassis.

A lot of electrical devices generate power line noise. This can get into an amp. I haven't worked this out for this amp but being on the primary side of the power transformer, I suspect that the line filter capacitor would act with the input impedance of transformer to create a low pass filter.

In terms of limiting the current on the chassis, the reactance of the capacitor = 1/ (2 * PI * F * C). The safety capacitor is 0.047uF. At 60Hz, the reactance is 1/( 2* 3.14 * 60 * 0.000000047) which is about 56K ohms. Assuming the line voltage is 125 VAC, using ohms law, I= V/R = 125/56,438 = 2.2mA. If the capacitor is connected between the hot power line and chassis ground, there can be up to 2.2mA at 125VAC on the chassis. Using a safety cap helps avoid the capacitor shorting as it age, if it fails it will open not short. If the capacitor were to short, you don't want 125VAC @15 amps on the chassis.


In the past, there wasn't as many sources of interference that would create power line noise. The filter capacitor worked fairly well. A lot more thought goes into line filters today. Some modern amps and power strips have EMI (electromagnet interference) filters built in. They are good to have and much better than what was used in vintage amps. These are much more comprehensive in the way that they filter noise present on the line power, they provide both differential and common mode interference filtering. The attached document provides an overview of the differential and common mode filtering.
 

Attachments

Last edited:
#70
The reason why the cap is not a filter (at least in any definitions that I am aware of) is because the filter would be part of a shunt element and since the chassis is not grounded there's no shunt element possibility.

It is a way to reference the chassis to something, in this case either conductor of the power line, one of which is likely to have an effective lower impedance to earth ground. Because the plugs back then were not polarized, there was no way to predict which side was going to be the grounded side (the ground bond was usually at either the service switchgear, or possibly at the pole transformer where the neutral is bonded to the transformer case and then to earth. This is done primarily for lightning protection of the grid and the load (mostly house structure). The cap value was chosen to limit the leakage current if the hot side was connected to the chassis.

The history of the development of the power grid over the last 100+ years is fascinating stuff, I have a whole shelf of books about it. Not only is there technology, but politics and economics involved.
 
  • Like
Reactions: webmonster
#71
Had all the tubes tested on a recently calibrated Knight tester; all checked out good, with no shorts. In fact, the original Philips/ECG 6L6 (the ones that had some sparking) tested virtually new, well into NOS range. Rectifier was great too.

So, I'm going to retension those octal sockets a bit and replace that cathode resistor with a modern wirewound version, just in case. I'll also clean the eyelet board and reflow some solder joints. Pretty sure that'll take care on my B-15 issues. Especially since all the voltages looked right at spec, or very close.
 
#72
Tube testers do not test for flash-over weaknesses, the specific fault that you have experienced.

It's like saying your water pipe has no leaks without pressure-testing.
 
  • Like
Reactions: RoadRanger
#73
agedhorse said:
Tube testers do not test for flash-over weaknesses, the specific fault that you have experienced.

It's like saying your water pipe has no leaks without pressure-testing.
Click to expand...

True enough. What I'll probably do is run a pair of new (less valuable) 6L6GC in there, after I do the above mentioned tasks, for a while. Trying to go for the most obvious culprits first. While I know it's not always true, I've found that often, it's the most obvious issues that end up being the problem. If the cheaper 6L6 pair runs fine for an extended period, I'll put my nice ones back in.

Turns out the pair of 6L6GC that were in there were some (supposedly) "holy-grail" tubes (Philips/EGC STR387, dual-getter)... I take that with a grain of salt, but they should at least be treated with some respect. I'd feel bad burning up any nice old tubes thru neglect.
 
#74
Frederick34 said:
True enough. What I'll probably do is run a pair of new (less valuable) 6L6GC in there, after I do the above mentioned tasks, for a while. Trying to go for the most obvious culprits first. While I know it's not always true, I've found that often, it's the most obvious issues that end up being the problem. If the cheaper 6L6 pair runs fine for an extended period, I'll put my nice ones back in.

Turns out the pair of 6L6GC that were in there were some (supposedly) "holy-grail" tubes (Philips/EGC STR387, dual-getter)... I take that with a grain of salt, but they should at least be treated with some respect. I'd feel bad burning up any nice old tubes thru neglect.
Click to expand...
I agree with you there. But I'll never buy another NOS 6L6 again. At one time I wasn't convinced of new 6L6 quality, but apparently they've gotten much better over the last few years because they're not going bad like previous new 6L6's I've tried. Will definitely stick with NOS 6SL7's, though...in a B-15 circuit, new ones suck. They sound great in the Jule Monique, but in a B-15, rather glassy and bleh.
 
#76
Final results...

Okay, just wanted to come back and post my findings/results for anyone in the future with similar issues:

1) I cleaned up the eyelet board & components the best I could.

2) I replaced the 1k/10w cathode resistor and the companion cathode cap. I did find that the original resistor measured 958k, so it really wasn't that far out of spec (this has been my experience in over 40 years of messing with electronics, I've never found an old resistor out-of-spec... go figure).

3) All the big caps had already been replaced (by me) before this sparking in power tubes began.

4) I added my 3-prong plug and disconnected "deathcap".

5) I retensioned (and had already cleaned) my 6L6GC octal sockets (I believe this was the real 'fix').

6) I ran some newer (cheap) 6L6GC for a while, at various volume levels and loads with perfect results.

7) I then put in the original "sparking" power tubes... they ran fine. Different loads/volumes. They had also tested perfectly on a calibrated Knight tester.

8) I then put in my nice pair of matched NOS Sylvania 6L6GC and the amp sounds phenomenal, couldn't be happier.

I am fairly certain now that the sparking was created by the octal power tube sockets being loose and/or some remaining 'wet' DeOxit in those sockets at that time. The reason I lean toward loose octal sockets is that the tube that sparked slid in and out too easily, and I believe I had heard that audible popping once or twice, after I'd replaced the big caps but before I used DeOxit on the sockets.

As far as resistors, how often has anyone actually come across resistors that have drifted out of spec? I literally never have. I replace them anyway, but I'm beginning to wonder if I should ever check that again. Granted, I'm not a pro amp technician, and I've probably only checked and replaced 20 or 30 resistors in my travels... but I've yet to find a bad one.
 
Last edited:
  • Like
Reactions: coreyfyfe and beans-on-toast
#77
Good to hear that it is resolved.

Resistors can go out of spec.

Mostly this happens with higher wattage power resistors that run hot. In cathode biased amps, check the cathode resistor. Check any higher wattage rated resistors such as those between power supply nodes. In amps with a dropping resistors in the bias circuit, check that one. Check plate and screen resistors if present.

Wire wound resistors can become microphonic.

How you test a resistor is also important. If it measures out of spec when it is cold, change it. But readings can drift relative to the nominal vaule depending on the temperature. Testing at the operating temperature can help in some cases.

In addition, carbon composition resistors can absorb moisture which changes how they read and their associated noise. Manufactures recommend that resistors be baked: http://www.rcdcomponents.com/rcd/press/Resistor FAQ.doc.
 
#78
Frederick34 said:
As far as resistors, how often has anyone actually come across resistors that have drifted out of spec? I literally never have. I replace them anyway, but I'm beginning to wonder if I should ever check that again. Granted, I'm not a pro amp technician, and I've probably only checked and replaced 20 or 30 resistors in my travels... but I've yet to find a bad one.
Click to expand...

Plenty, but I usually buy and repair old used/abused amps so maybe less of a surprise there. As @beans-on-toast mentioned, dropping resistors in the power supply, plate resistors, bias circuit, etc. I've replaced plenty of noisy carbon comps that tested okay. Depends on how they're rated but on my own amps if I'm digging through anyways I'll replace anything that tests off when cold.

Sometimes it's even easy to spot the bad ones -

18839119_10103557749343609_960051903620532552_n.jpg
 
  • Like
Reactions: beans-on-toast
#79
Not trying to be smart here but the consistent use of “6L6” as the output tube concerns me. The tubes need to be 6L6GC. The letters are important! If you put a 6L6 into that amp it would melt. Same for the B, GT, WGB ( industrial version of the 5881). These cannot take the B+ voltage in these amps. The tubes I threw out were JJ 6L6GC. They didn’t handle the the B+ well either in spite of being purchased for my B15.

I should add the the mains voltage in my home is typically 127 volts. Thus the voltages out of the secondaries are higher than what one would expect.
 
#80
BassmanPaul said:
Not trying to be smart here but the consistent use of “6L6” as the output tube concerns me. The tubes need to be 6L6GC. The letters are important! If you put a 6L6 into that amp it would melt. Same for the B, GT, WGB ( industrial version of the 5881). These cannot take the B+ voltage in these amps. The tubes I threw out were JJ 6L6GC. They didn’t handle the the B+ well either in spite of being purchased for my B15.

I should add the the mains voltage in my home is typically 127 volts. Thus the voltages out of the secondaries are higher than what one would expect.
Click to expand...
Really? Mine sounds fine with JJ's. I've tried them vs the RCA gray plates I used to have and they sounded the same to me. Think I have a GE set in there now, but honestly they all sounded the same.
 
You must log in or register to reply here.
Top Bottom
Back