Strange issue, debugging help

General discussion area for tube amps.

Moderators: pompeiisneaks, Colossal

User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Strange issue, debugging help

Post by WRC34 »

Hello all,

Back again after a long and frustrating process trying to debug a recent build. The issue is "crackle" or just ugly sounding distortion following notes when the amp is driven hard. A palm muted low E power chord or even worse a D power chord will produce a crackling that follows then dies shortly afterwards.

The amp is cathode biased EL34 with a 6SL7 ltpi and 6SJ7 input. The first brain buster was that the amp seemed to only exhibit the issue when standing straight up (either on top of the cab it was driving or not - and yes I tried having the head set up across the room from the speaker cab using a long speaker cable, and yes I tried different cabs/speakers all with the same results) When the amp was out of it's box and upside down on the bench it behaved perfectly. Put it back in the box and set it upright, any scenario, and crackle returned.

After re-wiring the output jacks, trying multiple screen resistor scenarios (started with shared 470 ohm 5W cement resistor, replaced with another of exact value and type, then switched to separate 1.5K 5W screen resistors, no change, and little effect on tone which I was initially worried about) I finally looked at the coupling caps. The first coupling cap which came off of the plate of the 6SJ7 and fed into the volume control (which then met another coupling cap going into the "top" half grid of the phase inverter) was a .022 and I was getting a 25mV reading on the phase inverter grids, so I replaced with a .03 PIO Russian type cap I had on hand. This completely solved the issue, but treble response was lacking. So, I replaced the .03 cap with a brand new .022 poly cap. The issue still seemed resolved but then in an extended jam session it returned. Luckily I had a backup pair of EL34s on hand which, when installed, seemed to solve the issue. However, in the real world working setting I found the amp still lacking in high end, so I replaced the .022 poly coupling cap with a Russian PIO .015 and the crackle returned. Another leaky cap, fresh off the shelf!? I suppose it's possible, so I installed an NOS Aerovox Duranite .0047 I happened to have on hand. Initial testing showed great results, high end response was dramatically improved and crackle issue still had not returned. I only tested the amp for a couple of minutes with the new cap, but it was at high volume and I put it through it's paces, everything seemed to be working fine.

I left town for a few days and came home to see the neighbors were gone. Taking advantage of the situation I attempted to test the amp more extensively and was shocked to find that immediately upon firing up and hitting the first note the crackle was back. Could the cap have gone leaky while I was away, after sitting dormant for decades then being put into use? It seems weird that this could happen, but I'm not ruling it out.

I have done the "leaky cap test" of removing output tubes and measuring DC voltage at the grids and the reading for both tubes is dead zero. I did the same test on both the 6SL7 phase inverter - removed it and measured DC voltage at both input grids - again 0 VDC. Finally removed the 6SJ7 and read the DC voltage at the input grid and again 0 VDC. And yes I have tried numerous different 6SJ7s, numerous 6SL7s, a few different pairs of EL34s and even swapped out the 5AR4 rectifier for another 5AR4. The only time this alleviated the issue was at the live jam when the newer (lesser used) EL34s seemed to solve the problem. Otherwise, no difference with any tubes.

I understand that this could be a parasitic oscillation, but then why would it go away then return? The amp behaves and sounds excellent otherwise, fairly quiet and great tone. Am I missing something?? I must be!
User avatar
Phil_S
Posts: 5936
Joined: Tue Oct 23, 2007 10:12 pm
Location: Baltimore, MD

Re: Strange issue, debugging help

Post by Phil_S »

What you describe sounds intermittent. Continuing to change caps is unlikely to solve the problem. Likely none of the caps were bad.
Look for a cold solder? Chopstick will help.
Change the other tubes. Power tubes have been changed.
These are simple things to do. Very little downside if nothing turns up.
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Strange issue, debugging help

Post by WRC34 »

Thank you Phil.

Both preamp tubes have been changed as well (all are known to work well in other amps). Chopsticking has shown no discernible problem zone or noise. At a total loss here. At one point I suspected a short in the output transformer core, but when the problem went away the amp sounded so loud & strong that it's hard to maintain that theory.
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Strange issue, debugging help

Post by WRC34 »

With new screen resistor setup output tubes are dissipating around 23.5 watts each. They were dissipating 24.5 with the shared 470R screen resistor - am I still too hot, and could biasing this hot cause an issue like this? I ask because the problem was resolved but resurfaced after an hour or so of playing loud w/full band. A swap of output tubes to a "less tired" set seemed to solve the problem. The first pair of output tubes are Yugoslavian Ei 6CA7s, and the "new" set are Ultron branded RFT EL34s which were used for another 30 mins or so. Both sets of tubes had been used previously. The 6CA7s had more hours on them, but both only used in fixed bias amps and biased to a max 70% at idle and most of the time more like 60 - 65%. Could this max dissipation be creating a noise issue in the tubes this quickly?

I may be grabbing at straws but I am desperate at this point!
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Strange issue, debugging help

Post by WRC34 »

Scratch that possibility, just tried a pair of ANOS Siemens EL34s, no change. And now I'm ruling out the coupling cap (at least the first one) because I just swapped out ANOTHER cap (.01) and also no change.

Any suggestions are welcome and appreciated
frankdrebin
Posts: 131
Joined: Wed May 14, 2014 11:40 am

Re: Strange issue, debugging help

Post by frankdrebin »

it could be arcing inside a tube or in a socket,try looking at those in the dark while playing hard.
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Strange issue, debugging help

Post by WRC34 »

frankdrebin, thank you for your suggestion!

I have looked very closely at the sockets/wiring while playing and have not seen any visible arcing. Is it possible that a tube socket could be bad and cause this issue on it's own?
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Strange issue, debugging help

Post by WRC34 »

Another possibility just occurred to me...

Has anyone had any issues with brand new F&T capacitors recently? A year ago I built a fixed bias 6V6 amp for a friend and after some troubleshooting I discovered a bad filter cap (an F&T 22uf 500V which fed the phase inverter). The DC voltage reading at idle was jumping around all over the place. I don't remember the audio sign that told me there was a problem but it took me a while to track down since I did not suspect a brand new cap to be bad.

DC voltage readings seem stable at idle in this build, but perhaps one of the filter caps is bad. The only crappy thing about this revelation is to replace each filer cap with another new or "good" cap one at a time and test for success, since I don't have an ESR meter or anything like that. And if I'm wrong, well, it'll be a bunch of work for nothing but at this point I have pretty much stopped counting. The amp sounded so good when it was working properly that I don't care about man hours I just want it to be operational!

If I'm right though, it makes sense that when the thought occurred to me I felt like the Dude when Maude tells him her father doesn't actually have any money ;)
Firestorm
Posts: 3033
Joined: Fri Jan 25, 2008 7:34 pm
Location: Connecticut

Re: Strange issue, debugging help

Post by Firestorm »

What are the plate and screen voltages on the EL34s?
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Strange issue, debugging help

Post by WRC34 »

All readings are DC voltage

Plate - 340
Screen - 323
EL34 cathode to ground - 20
Plate to cathode - 320
Plate supply (OT CT) - 345
Screen supply (filter cap that feeds EL34 screens) - 340
Cathode resistor (one for each output tube) - actual measurement 265R each but rated for 270R. These are 10 watt resistors, each bypassed with their own 220uf 100V cap

Output transformer is 6.6K, using 8ohm secondary tap into various 8ohm cabs (not at the same time, but I have used 3 different cabs for testing t make sure the issue is not cab-related)

I am now confused and cannot tell if my combined screen and plate dissipation is over spec or under. The cathode bias resistor values are in line with some classic EL34 cathode biased torture chambers, most notable Matchless Clubman, but plate & screen voltages are very conservative. Still, I'm either right below or slightly above 25w dissipation per tube, which would explain why/how this amp sounds so strong. It sounds louder and stronger than any JTM45 clone I have owned (3 in total). There is no negative feedback used in the amp. The OT is a 35 watt rated old timey Northlake that I got NOS but used in a 5E3 build prior (in that amp it was well oversized/over-rated of course) but I'm almost entertaining the chance that I may have slightly stressed/fried the OT by running it in such a powerful amp.

FWIW I have not tried swapping any of the filter caps yet.
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Strange issue, debugging help

Post by WRC34 »

Maybe I can somehow post a soundclip or video, maybe that would help explain what's happening.

I have posted like 100 times on my own thread with very few replies. I know this isn't cool but I don't care - been wrestling with this problem for weeks now and I want to fix it/learn what is wrong.

Especially frustrating is the fact that twice it has seemed to be fixed and the reverted back to being f*ked, one time from playing and the other time from simply sitting unused for 3 days. WTF!?
frankdrebin
Posts: 131
Joined: Wed May 14, 2014 11:40 am

Re: Strange issue, debugging help

Post by frankdrebin »

Apart from adding schematic, I suggest getting an old crt scope, that will save thousands of hours in debugging.
I have a guy that has brought me 6 amps built by him just because he thinks its easy, but everyone showed different problem and meters not always tell you what is going on.

Inviato dal mio m3 note utilizzando Tapatalk

User avatar
Malcolm Irving
Posts: 139
Joined: Wed Jan 14, 2015 12:06 am
Location: Derbyshire, UK

Re: Strange issue, debugging help

Post by Malcolm Irving »

If you are competent to work on a live chassis, have you tried this:
With the amp powered up and a speaker connected, but no guitar, put all the volume controls right up.
Give every component, connection, solder joint, wire and socket a few good hard taps with a chop-stick. (Just use one hand and keep the other in your pocket.)
You will find a lot of the components are slightly micro-phonic – that’s not a problem. But hopefully one tap somewhere will make the crackle and you can identify the joint/component which is responsible.
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Strange issue, debugging help

Post by WRC34 »

I don't consider amp building to be easy! I enjoy the challenge and have built around 20 amplifiers to date. Normally, upon completion, there's some minor troubleshooting to do. Sometimes an amp will work flawlessly in the first start-up. I don't have a scope yet but I do hope to get one in the near future! This issue has been particularly confusing, hence my appeal for help here. In the past the members here have been really helpful and I have always appreciated that.

I have chopstick'd the amp a few times, hoping to turn up the issue as simply being a cold solder joint. This was discussed earlier in the thread. And I consider myself more than competent to work on a live chassis - how else would I have obtained the voltage readings I posted?
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Strange issue, debugging help

Post by WRC34 »

And hey, I apologize if I seem snarky in response guys, I am grateful for ALL replies here - I have a 2 year old crawling on me like a jungle gym as I type, and I feel like I have taken pains to give detailed info rather than ask vague questions!
Post Reply