Hum in Tube-Driven Reverb

[mistersparkle]

Greetings, all.  Been a member here for five years but never had a reason to post. Just hanging out, fat, dumb and happy, building my low-voltage, solid-state effects, until I took on this high-voltage challenge. Now, this isn't really about a "stompbox" per-se, but it is about a guitar effect unit.

I scratch-built a clone of a Fender 6G15 stand-alone, tube-driven spring reverb, right down to the finger-jointed, Tolex-covered pine cabinet, and I must say it works/sounds fantastic. But, I have one nagging issue that I haven't been able to crack. The unit hums when the "Dwell" control (aka reverb drive) is turned all the way down. Of course, this doesn't affect the actual  performance of the unit whatsoever, but it bugs me.

I've built tube amps from scratch previously, so I'm no stranger to the established techniques for eliminating ground loops (i.e. isolated input and output jacks with a single-point grounding bus bar, etc.), but all my attempts to bust this hum have been unsuccessful. If there are any tube experts out there with any ideas, please let me know. I do have copious pix and other docs so please let me know if I can post anything that might help. I included a pic of the control panel just for giggles.

PS: Your verification questions are jacked up.  It's not "An ukelele" that has 4 strings, it's "A ukelele", and it's not an "RC4588" that has 8 pins, it's an "RC4558", dummies, begging the question, "Do I trust any answers I might get from this forum?"

[tubegeek]

You've misused the phrase "begging the question," so how can I trust any questions you might ask about the problem with your Reverb unit?

I'll try to get past that. 

Is it possible the power transformer is magnetically inducing the hum in the reverb return transducer? How close together are they? Or perhaps AC heater wires too close to that same transducer?

You may want to put a schematic in your post. Or else somebody else will ask you to.

Welcome. Enjoy.

[mistersparkle]

I know, I know, I technically misused "begging the question", but only Marilyn Vos Savant and Wikipedia have issues with the colloquial use of the phrase. And not really thinking that know-it-all woman would be reading my post, I took the chance. Maybe you're related to her, but I can get past that. 

I appreciate your suggestions, tubegeek, but I don't think either of them are the problem. As I noted, the unit works perfectly with no hum whatsoever until I turn the Dwell control all the way down - something you would not normally do when using the unit. With that control turned down I get hum in all positions of the Mix control - full wet, full dry or anywhere in between. Given the control's position in the circuit, it just doesn't make sense.

I've attached the schematic my build was based on - Ted Weber's 5G15 clone kit. It's identical to the Fender original with the exception of the power supply design (which I liked better than the original), and the use of a modern 6V6 power tube to drive the tank instead of the now-obsolete 6K6 Fender used. I deviated from the Weber schematic by using a 6K6 and changing the output transformer to one from Triode electronics that matches that tube.

[willienillie]

Have you tried a different 12AT7?

How is your Dwell pot grounded?

[willienillie]

Also, just noticed the Weber schematic shows a 3-prong AC cord.  This can/will create a ground loop when connected to your amp.  I would at least add a "ground lift" switch.  You can easily test if this is related to your hum problem by using a ground lift plug adapater...thingie.

[PRR]

Grounding scheme. Clearly (as willie says) the Dwell pot return goes to a dirty ground. You say you know this stuff but have not offered any details we can look over to make suggestions.

[tubegeek]

I was hoping maybe the hum was there but masked by some signal when the dwell was up. Seems like not. Better suggestions from others anyway. Let us know how you fare?

[mistersparkle]

Thanks for all the suggestions. I know I haven't been able to provide the level of detail you'd all like, but I'm currently far from home on vacation with time on my hands to catch up on my various forums. When I return I'll try the AC ground lift idea, swap the 12AT7 and also quadruple-check the grounding on the Dwell pot.

As PRR states, it's clearly a grounding issue. But even with my 40+ years of experience chasing ground loops in my professional life (commercial audio/video systems design engineer), this one has me flummoxed. Perhaps I should break a grounding rule or two on this one - it wouldn't be the first time I busted a hum doing what would technically be considered unorthodox. I've attached a pic that shows the basic grounding scheme - bus bar running the length of the turret board I used, with pots and jacks returning to the bus. I could try returning each pot & jack directly to the bus, like I do in amp builds, but this chassis is so small I didn't deem it necessary.

[PRR]

A map would help. The left blue cap is main B+ filter? The left pot is the reverb? Dot yellow line is its return?


IAC, that pot and a jack *appear* to return to the *worst* part of "ground". The red barb-wire line is the path from PT through rectifier to first cap. This path is *full of crap*, the 100/120Hz pulsations from PT to filter. Reverb send/return should probably return to the *other* end of the ground wire, the quiet end.

Don't drill your well under your outhouse. Don't make your kids play in the through-street with the heavy trucks.

[mistersparkle]

You are correct in your observations about the layout and make a good point. Perhaps the problem could be remedied by taking the pot/jack ground lead directly to the single-point chassis ground rather than to the turret board ground bus. In fact, maybe the bridge, turret board bus and pots/jacks should all return to the single ground point. I'll try messing with that when I get back home.

This is my first experience with a SS rectifier in a tube circuit so I may need to learn some new rules. Fwiw, early construction stage photo attached showing the single-point ground lugs for filament winding CT, AC line ground (not yet installed - not sure what that 2nd green wire is) and grounds for the isolated RCA tank jacks. Looking at it, I'm thinking I did the back panel correctly and messed up on the front panel.

[PRR]

> taking the pot/jack ground lead directly to the single-point chassis ground

WHY ? ? ? ?

(Chassis is not a magic garbage dump.)

It should return to the same point as the cathode network of the tube taking signal from the wiper.

[anotherjim]

I think this article provided by Valve Wizard (Merlin) will save me a lot of typing...
http://www.valvewizard.co.uk/Grounding.pdf

[mistersparkle]

Well, given that my favorite programming language is solder, I'll heat up the iron next week, disconnect the Dwell pot ground and clip-lead it until I find the magic clean ground it's looking for. Ironically, I won't be able to do that until after I fix a friend's ultralinear Fender Twin that......uh, hums. But in this amp's case it's clearly a filter cap issue, nothing spooky that I'll have to post about in an amp forum.

Thanks again for all the suggestions, and I will post the results of my "reprogramming" efforts when I get the time to tackle it.

[amptramp]

There is one additional thing to try.

Some 12AX7 constructions let the plate "see" the filament because under the mica spacer, the wires run out but also on the top, in some cases there is a pathway for the filament to emit electrons directly to the plate.  The cure for this is to take the centre tap of the filament ground to a positive voltage like the 6V6 cathode because the cathode follower for the input signal is already at a high cathode voltage and this will reduce some heater to cathode voltage stress.  With the filament at a higher average voltage, the grid wires that are in the way will be much more negative than the filament and will block heater emission.

[mistersparkle]

Yes, amptramp, raising the heater winding CT above ground is something I'm aware of but admittedly have not yet tried in this case. I'll definitely add that to my list of things to check out once I'm able. I learned a loooong time ago, from several Heathkit manuals, that a human will overlook the same error repeatedly while a fresh pair of eyes will spot the trouble immediately (ask my first Corvette motor teardown/rebuild how I know this to be true). And again, I'm thankful for all the fresh ideas offered up here, some more congenially than others (which is expected in most forums) and I will post my results as time permits.

[tubegeek]

Don't mistake demented sarcasm, edited down too far (because of thumb-typing) for a lack of congeniality!

And - if you weren't referring to me, I'm insulted!

[mistersparkle]

Don't mistake demented sarcasm, edited down too far (because of thumb-typing) for a lack of congeniality!

And - if you weren't referring to me, I'm insulted!

I would have been insulted had you not noticed my demented sarcasm. And now that I finally got my avatar to display properly, between that and my signature you should be able to tell that I'm 99.9% humor and .1% serious. Sometimes an errant streak of intelligence might sneak through, but I do my best to suppress it. Cheers!

[tubegeek]

Sometimes an errant streak of intelligence might sneak through

Some of us do NOT have to worry about that.

[tubegeek]

(Chassis is not a magic garbage dump.)

I wish I wasn't so happy already with my current sig block. Very tempted to make a change. Might go with a dual quote in fact.

EDIT: Couldn't resist.

[mistersparkle]

Well, after returning to the US following a few weeks of vacation, then getting all wrapped up in holiday stuff, I finally dug out the reverb unit to see if any of the offered suggestions (including getting rid of that "magic garbage dump" of a chassis ground) would solve my nagging problem. I suppose I owe you folks an update, and so........the winner is...........wait for it..........nobody (sigh).

Believe it or not, I could not reproduce the problem. I tried the reverb with four different amps, two tube & two solid-state (including the Vox Pathfinder I was originally using when I discovered the problem), and no hum. I will offer that the quiescent hum is just a tad lower with the AC ground lifted (thanks for making me try that, willienillie), but not enough to compromise the safety factor of that ground.

Sorry for the wild goose chase, folks. I hope at least it was as much fun for y'all as it was for me. I suppose I'll go ahead now and finish the second unit I promised to a friend. Anybody else want one?? I can make more....