Debuzzing a valve reverb circuit...

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

robertbonello

Member
Joined
Dec 14, 2011
Messages
15
Hello all.

I'm very much at the beginning of getting to grips with stuff and have something I've come across which I can't seem to find the answer for. I've got quite a few successful project builds under my belt, but I've recently decided to challenge myself a bit by building a simple tube reverb based on an old guitar magazine project - http://www.musikding.de/images/product_images/popup_images/2897_1.jpg that's less documented than most of the great projects you guys offer.

After very many beginner mistakes(but hey, I'm learning loads!) and lots and lots of reading, I've surprisingly got the thing working! Albeit not ideally...

The reverb sounds very dark, which is something I'd love to ask about and deal with later, but firstly, I've got a rather big problem with buzz coming though. I haven't got access to an oscilloscope so I'm trying to troubleshoot with a DMM and a dc blocking capacitor in series with a speaker as a "probe".

My power rails seem to be in good health. I've got the tube heaters being fed with AC. At the input of the first amplification stage I'm getting a 60mV or so DC reading. Should I read something into this?

Anyway, onto the buzz...

With an 8ohm speaker attached to the spring tank input cable, I'm able to hear what's going on, and the buzz is definitely present. Plugging a guitar lead into the input and monitoring the output, i did some very careful but uneducated poking around.

I found that when i place the tip of the input guitar lead touching the insulation of the black heater wire(from supply to tube), the noise drops substantially. When I touch the insulation of the red wire, it gets a bit worse!

Just to be clear, I'm not poking around and making any electrical contact anywhere in the circuit, I'm placing the tip of the guitar jack against the insulated cover of the heater wire. All heater wires are quite well twisted together, and even so, depending on whether the guitar jack is touching red or black , it makes a substantial difference.

Can anybody please explain/guess what's going on here, and how I might have screwed up? I guess I could cobble together a fix involving some enterprising wire wrapping, but that won't teach me anything about where I'm going wrong.

Unfortunately, I've built the project in a very silly case with the terminal strips at the bottom of the case and the tubes, transformers,choke, jacks and pots on top, so it's not something I can take a photo of easily. I've tried spreading and bunching the various wires(with a wooden spoon) but there is no change in the noise. The only thing that's made a difference has been the guitar lead tip to the insulation of the heater wire as described above.

Because of my immensely stupid case "design", my grounding scheme has also been a bit all over the place and I've been trying various variations of this. My guess is that I'm doing something very dumb here. I've since collected all my grounds and joined them together, along with the case and everything else. My power transformer centre taps go to the ground rail of my PSU terminal strip, which then joins up to star ground, along with all other grounding wires from audio ground, in & out jacks, case, pots etc

Photo attached to show my genius case setup taken from the back...

Any guidance would be massively appreciated.

Thanks, Rob
 

Attachments

  • photo-8.JPG
    photo-8.JPG
    456.6 KB · Views: 40
bolt all the grounds to chassis real good,

get a .1 uf/400 volt cap and clip one end to ground with a wire,

put the cap lead that is not grounded on the grids of all the tubes,

see if there is a drop in humm,

make sure al the tubes have the same heater wire going to pins 9, or whatever,

in other words, don't cross your heater wires up, organize them so 1 green wire goes to all the pin 9's, or whatever pin is one of the heater pins,

then connect the other green wire to pins 4/5 if using 12ax series,

make sure pwr trans is away from reverb tank,



 
Hi CJ,

CJ said:
bolt all the grounds to chassis real good,

CHECK

get a .1 uf/400 volt cap and clip one end to ground with a wire,

put the cap lead that is not grounded on the grids of all the tubes,

see if there is a drop in humm,

ONLY A SLIGHT DROP ON v2 pin 7, BUT NOWHERE NEAR THE BUZZ DROP I GET WITH THE GUITAR LEAD TO INSULATION TRICK

make sure al the tubes have the same heater wire going to pins 9, or whatever,

in other words, don't cross your heater wires up, organize them so 1 green wire goes to all the pin 9's, or whatever pin is one of the heater pins,

then connect the other green wire to pins 4/5 if using 12ax series,

NOT SURE IF I GET YOU HERE. BOTH TUBES ARE HEATED WITH PINS 4&5. SAME TAP FEEDS 4 ON BOTH AND OTHER TAP FEEDS 5 ON BOTH. TUBES ARE V1: pcl86 (14gw8??) AND V2: ecc83(12ax7)


make sure pwr trans is away from reverb tank,

THERE'S A GOOD 6 INCHES BETWEEN THE 2, WITH THE GROUNDED CASE IN BETWEEN. ALSO, THIS SHOULDN'T COME INTO PLAY AT THIS POINT. I'VE HOOKED A SPEAKER UP TO THE CABLE THAT FEEDS INTO THE REVERB TANK, SO I'M MONITORING THE SIGNAL PATH UP UNTIL THE OUTPUT OF THE TRANSFORMER. AGAIN, NOT SURE IT'S THE THING TO DO BUT I'M TRYING TO ISOLATE THE PROBLEM HERE.

any thoughts? Thanks a million CJ, much appreciated
 
ruffrecords said:
You do have the heater supply grounded via the pair of resistors shown in the psu schematic?

Cheers

Ian

Hi Ian,

I do actually have them on the terminal strips as I had trouble figuring out some fundamentals with the PSU earlier on, but I've lifted them. They are not in the v2 power supply schematic. The different versions of the power supply are for different power transformer setups. My transformer has 2x 6.3v secondaries with CT to psu ground. Connected in series for 13.5vac feeding the valves.

Cheers,

Rob
 
robertbonello said:
ruffrecords said:
You do have the heater supply grounded via the pair of resistors shown in the psu schematic?

Cheers

Ian

Hi Ian,

I do actually have them on the terminal strips as I had trouble figuring out some fundamentals with the PSU earlier on, but I've lifted them. They are not in the v2 power supply schematic. The different versions of the power supply are for different power transformer setups. My transformer has 2x 6.3v secondaries with CT to psu ground. Connected in series for 12.5 - 13.5vac feeding the valves.

Cheers,

Rob

OK, if you are doing as it says in the V2 scheme then that should be OK. The only other thing I can think of that might be causing this is that there is no decoupling of the cathode resistor of the triode of the PCL86 - R3. This is a common technique that new designers think reduces distortion by negative feedback but as often as not it does the opposite by increasing ra. As a side effect it makes the stage more susceptible to heater induced hum. Try decoupling it with a 100uF 25V cap and see if that helps.

Cheers

Ian
 
Depending, could possibly be your audio transformers inducing hum into the circuit from the power transformer and power and the EM fields all around.
 
I think it is time for you to invest in a scope. Right now we don't know if your buzz is 60Hz or 120Hz which tells us a lot about the probable causes.

Cheers

Ian
 
ruffrecords said:
OK, if you are doing as it says in the V2 scheme then that should be OK. The only other thing I can think of that might be causing this is that there is no decoupling of the cathode resistor of the triode of the PCL86 - R3. This is a common technique that new designers think reduces distortion by negative feedback but as often as not it does the opposite by increasing ra. As a side effect it makes the stage more susceptible to heater induced hum. Try decoupling it with a 100uF 25V cap and see if that helps.

Cheers

Ian

Hi Ian,

I tried a 100uF 25v cap in parallel(that's correct yes?) with R3. Made no difference I'm afraid..

I've been looking at oscilloscopes for a while, I really do feel I need one, but would love to find a decent second hand one(as I'd use it for very little). I'm in Malta so there's not too much of a second hand market for things like that. I'm sniffing around. Nevertheless after checking I wasn't going to cook anything, I hooked the rca reverb IN to my DAW and put it through what frequency analyser I had. Screenshot attached.

I'm reading the peaks at:

53hz at -58.8
106hz at -65.8
152hz at -47.9
199hz at -63.8
245hz at -49.1
351hz at -50.6
447hz at -68.0
539hz at -63.8
635hz at -73.4

Noise from 1k to 4k is in the -60 to -70 range
3.177khz loudest peak at -58.9

Levels were not calibrated to anything, so all db values are merely given as a point of reference to what I'm guessing should be the noise floor between 1khz and 4khz

Does this help at all?

Thanks a bunch,

Rob
 

Attachments

  • Picture 3.png
    Picture 3.png
    19.3 KB · Views: 7
I took a look at the "guitar jack to heater winding insulation" trick.

It shows no difference to my 50hz, 100hz etc etc hum, but drops my 1-6khz noise by up to around 20db(again, of this uncalibrated scale)

I'm in way over my head at this point. Any suggestions or even directions I might be pointed in would be really appreciated.

Screenshot attached

Thanks,

Rob
 

Attachments

  • Picture 4.png
    Picture 4.png
    20 KB · Views: 7
disable each tube stage one at a time,

start at the back,

work your way forward,

you can not troubleshoot an entire schematic,so you break it down into smaller sections, and then smaller sections til you find the problem,
 
+1 to what CJ said. You need to pin point where it starts. Where are you measuring for the graphs you posted and what are the conditions?  e.g. what is plugged into the input?

Edit. The first thing to do is short the input so you know its not getting in there and measure again.

Cheers

Ian
 
Hi Rob, I think at least your Tube Reverb is oscillating! There was some information about that issue on the former website of the designer, who died last year! I have this Tube Reverb and had some problems with oscillating, too. But now everything sounds really good! I have fixed that issues without the "cure" you can find in the attachment.

Please let somebody experienced check your build on safety issues, because there were some basic problems in the early design/ description of this tube project. Take care for proper grounding and safety earth!!!

I am not responsible for the "cure" or anthing else, this is just my personal opinion!

Good luck!

rock
 

Attachments

  • image.jpg
    image.jpg
    84.8 KB · Views: 24
Are the coupling capacitors not a bit on the small side?
(10nF/250K, 680pF/680K)
But maybe it is the idea that they act as a high-pass filter.
You probably won't have too many 'boing' and 'bang' sound in your reverb...
 
If you move the reverb pan around does the buzz change? It should. Also, the send and return cables need to be shielded, but only one shield connected to the pan, to prevent a ground loop. Some pans one rca jack is isolated.
 
With an 8ohm speaker attached to the spring tank input cable, I'm able to hear what's going on, and the buzz is definitely present. Plugging a guitar lead into the input and monitoring the output, i did some very careful but uneducated poking around.

So are you saying that you are getting buzz with no guitar plugged in as well?  If you haven't checked that yet I would suggest doing so

  It looks like this design uses a pot on the first tube grid - turn wide open and then off and see if buzz changes (with no guitar plugged in).

With a guitar plugged in the input and *buzz* - you have to ask other questions - 1) single coil or humbucker.?  2) CRT computer monitor on in same room?  3) does rotating the guitar change the buzz? 

Also, make sure and tie the test speaker negative side to circuit ground as you would a normal amp speaker.  I've seen that cause more oscillation (squeal) than buzz but it could be a culprit for some noise in general if not tied in correctly.


I found that when i place the tip of the input guitar lead touching the insulation of the black heater wire(from supply to tube), the noise drops substantially. When I touch the insulation of the red wire, it gets a bit worse!

Could you please clarify here - how are you using the "tip of the input guitar lead"?    Are you using a guitar cable plugged into the input and then using the loose end as a probe?  Or are just jumpering from in-jack cable tip to other places?  I'm not clear on how you have things connected here and some of them are obvious ways to create big buzz.
 

Latest posts

Back
Top