Hum troubleshooting Dumbelator clone

[ElectronAvalanche]

Hi gang,

as reported in another thread I finished my Dumbleator clone.

I used two flatpack transformers (230v -> 9V -> 9V -230V) to get the B+ which of course is a lot lower than in the real deal (195V on the CF, 21V cathode and 126V on the anode of the return stage).

There is a tiny bit of 50hz hum on the recovery stage (I think)

grounding the circuit after the 0.1uF kills the hum,

grounding at any point before that does not have any influence on the hum.

Thus when having the recover output pot all the way down: no hum.

When turning off the mains of the Dumbleator the hum is gone, the signal is still there until the filament is cold for no electrons to be emitted and the B+ is drained.

The heater voltage is derived from the 9V after the first flatpack into a FWB into 7806 regulator and caps + diode (plain vanilla voltage regulator circuit). I think the heater voltage must be th problem then.

I do not have a scope at hand at the moment, so I cannot see if the heater voltage has some 50hz ripple on it. I have 5.91V DC on the heater.

Any ideas how to fix this? What could cause the 50hz hum?

I have already ordered a toroid transformer that will give me 270V AC (80mA) and 14V AC@1A. This should give me the better volatges. I could then do a regular AC heater wiring with the ECC83 wired for 12.6V and some series resistance to drop the voltage.

Or I could make a 12V DC (regulated) supply. I figure the DC should give me a lower noise floor.

Cheers and thanks for any input!

Electron

[SoundPerf]

I would think you would need atleast some filter caps on the rectified 9V supply. When I build regulated effect PSU I use a 220uf on one side and a 100uf on the other side of the regulator.

[ElectronAvalanche]

Hi Chris,

I have one 680uF on each side of the regulator.

Electron

[glasman]

Try installing a 0.1uf disc ceramic on the plate of the Cathode follower to ground.


Also isolating the jack on the back and returning them to a star point really help (if you have alreay done this I appologize).

Gary

[glasman]

FWIW, I decouple all of my plate feeds in both my Matchbox and ReverbALoops to eliminate potential issues.

Gary

[glasman]

Also did you ground the fillaments common leg (pin 2 of 7806) to the AC ground. If not it will HUM like crazy.

Gary

[ElectronAvalanche]

Try installing a 0.1uf disc ceramic on the plate of the Cathode follower to ground.


Also isolating the jack on the back and returning them to a star point really help (if you have alreay done this I appologize).

Also did you ground the fillaments common leg (pin 2 of 7806) to the AC ground. If not it will HUM like crazy.

Gary

Hi Gary,

thanks for your reply and tips!

The 7806 ground leg (2) is grounded to the ground plane of the PS. By AC ground do you mean to the chassis ground?

In this Dumbleator I implemented a ground lift (which btw does not change the 50hz hum).

The hum I have is very low frequency, say pure 50hz sine-wace without any buzzing component to it. The level of the hum is not that bad, I can hear it and with the recovery maxed it is audiable but not bad.

With the 0.1uF ceramic going from the plate to ground you mean a cap that is essentially paralleled to the 33uF filter cap of the CF? What exactly is the function of the 0.1uF?

The jacks are all isolated from the chassis. There is sort of a star ground scheme going on with the ground-lift (ala TUT) thrown in.

Thanks again for your help!

Electron

[SoundPerf]

Hi Chris,

I have one 680uF on each side of the regulator.

Electron

Sorry about that...I was thinking you would know to do this. I was looking at the photo and I couldn't quite see the specifics.

[ElectronAvalanche]

Hi Chris,

no sweat! The pic is really not that good to see the details.

Really appreciate your help!

Odd thing is that I used this Power supply in a different enclosure as a supply for a Dumbleator for some years and then for a 2 tube preamp and can not remember that it had that hum.

When the new PT arrives I will make a new PCB for the Power supply. We will see if that helps.

Thanks again!

Electron

[bepone]

i think that ground loop occurs from grounding heaters supply 7806, and High voltage DC, which coming from the same AC source (if i understand correctly). one power transformer will resolve problem if so..

[Luthierwnc]

I put a bypass on these things. If you did, does it hum in bypass? If not, you might try unsoldering the jack hots and shorting them together to be sure the problem isn't in the cables.

Here's what the Power Supply looks like on my Frankenrouter. I'm pulling 12v DC from the original power supply so the toroidal is only the HV. Wired in series the secondary is supposed to be 240VDC but it is a little hotter -- which worked out well. The fan is disconnected.

Skip

[markusw]

Did you try different tubes?
Do you ground heater supply and HV at the same spot?
I have them separated. HV ps is grounded together with the isolated Neutrik jacks, pot bar and the "main" PCB.
Don't know if that matters.
Seems you have three HV ps caps? What values did you use?
Is it 50Hz or 100Hz you hear?

Regards,

Markus

[glasman]

Try installing a 0.1uf disc ceramic on the plate of the Cathode follower to ground.


Also isolating the jack on the back and returning them to a star point really help (if you have alreay done this I appologize).

Also did you ground the fillaments common leg (pin 2 of 7806) to the AC ground. If not it will HUM like crazy.

Gary

Hi Gary,

thanks for your reply and tips!

The 7806 ground leg (2) is grounded to the ground plane of the PS. By AC ground do you mean to the chassis ground?

In this Dumbleator I implemented a ground lift (which btw does not change the 50hz hum).

The hum I have is very low frequency, say pure 50hz sine-wace without any buzzing component to it. The level of the hum is not that bad, I can hear it and with the recovery maxed it is audiable but not bad.

With the 0.1uF ceramic going from the plate to ground you mean a cap that is essentially paralleled to the 33uF filter cap of the CF? What exactly is the function of the 0.1uF?

Electron

You are correct it is in parallel with the PS filter cap. It provides decoupling right at the point of use. More useful in solving a weird ground loop that what you are describing above.

Would suspect the hum maybe from the flatpacks. Some well grouded copper shielding might also help with the noise.

If I were going to build a fully contained d-lator again, I would opt for torioidal transfomers as they are inherent hum cancelling due to their design.

Gary

[Luthierwnc]

The last D'lator I made used an flatpak toroidal and it was well-behaved. The Amplimo line is a Dutch company and I think they make a pretty good pre-amp toroidal. The one in my picture is the cheapest Mouser PT they had but I didn't need a heater winding.

It is hard to troubleshoot something you can't hear or see and it also encourages lots of recommendations you have already tried -- but are the amp and the loop power-grounded at the same outlet? I've got a cheap power conditioner to help with that. We go through elaborate pains to keep the signal path away from ground loops but the AC lugs on the amp and the loop have the length of both cords and chassis distance to pick-up voltage potential. If I'd thought more about about it, I'd have built a convenience outlet on the amp and used a really short power cord to the loop.

FWIW, Skip

[ElectronAvalanche]

Just a quick follow-up:

I replaced the flatpack board with a toroid and a newly made PCB.

Low and behold, hum is gone, Loop is dead (!!!) silent and sounds great.

Voltages are 270V on the CF anode (380V B+) and 240V on the other system plate. 29V on the CF cathode.

Heaters are dead-smack 12.6V using a 14V AC from the toroid and a 7812 regulator with diode on leg 2 to ground.

Thanks for all the tips. In the end Gary was right, it was probably the flatpacks.

Take home message:

will use toroids in cramped 19" enclosures.

Oddly enough with the way lower voltages using the flatpacks, the Dumbleator had a tad more character. Maybe due to the CF clipping. But very musical.

Cheers,

Electron

P.S.: this time I used chemical "tinning" for the PCB and that worked out nicely. Using solder to tin the traces as on the main board is not recommended. I am very happy to be able to make PCBs DIY now. Safes a lot of time and makes for a very clean product.