Hum - can't find an answer by searching the forum

[Simpleton]

Hello everyone!

I am having some troubles with my DIY Paradise Simple EL84.
The amp sounds great, but I have hum problems, namely what seems to be 100Hz. Input cable from RCA to the pot is shielded and grounded on one side. The stereo inputs are joined by a passive mixer made with 2x 10K resistors.

The schematic is attached.
Version 1 is the one I find that cures the hum, which gets reduced to a very minimum level, almost inaudible. On the other hand, volume control seems tricky (I've tried with 3 different pots).

Version 2 has audible hum, which increases with increasing volume (zero hum audible at the first 1/16 turn of the pot and at max volume). If I put a 10K resistor across the wiper and the ground on Version 2, hum goes away, but so goes most of the sound.
Adding more capacitance to the PSU has no audible effect, and changing the sources (portable cdp) has no too.

This photo was taken some time ago, with an ECC88 as preamp tube, same story as of now. Just to show the component disposition on the plastic enclosure.

http://ngage101.no.sapo.pt/compact/IMAG0122.JPG


PSU is 1n4007 bridge rectifier, 440uF//100nF - 420 Ohm - 900uF//470nF.
The power tubes are biased at aprox 30mA each, preamp at around 10mA.

Thanks for any help given

[SY]

Photos of the wiring would be helpful. 100 Hz would suggest a grounding issue.

[Simpleton]

Photos:

http://ngage101.no.sapo.pt/amp/HPIM0237.JPG
http://ngage101.no.sapo.pt/amp/HPIM0238.JPG
http://ngage101.no.sapo.pt/amp/HPIM0239.JPG
http://ngage101.no.sapo.pt/amp/HPIM0240.JPG
http://ngage101.no.sapo.pt/amp/HPIM0241.JPG
http://ngage101.no.sapo.pt/amp/HPIM0242.JPG
http://ngage101.no.sapo.pt/amp/HPIM0243.JPG
http://ngage101.no.sapo.pt/amp/HPIM0244.JPG
http://ngage101.no.sapo.pt/amp/HPIM0245.JPG
http://ngage101.no.sapo.pt/amp/HPIM0246.JPG

Thanks Sy

[AndrewT]

Hi,
version 1 has DC passing to ground via the 100k volume pot.
Is that what is intended?

Similarly, version 2 grounds some of the grid current? and worse you're passing DC through the wiper?

Input cable from RCA to the pot is shielded and grounded on one side. The stereo inputs are joined by a passive mixer made with 2x 10K resistors.

you need two inputs from your source, a flow and a return. If the screen is connected at one end only you have broken the return. It appears the flow is returning via some unintended route and you're lucky it works at all.

[janneman]

Version 1 could work, when you place the pot AFTER the coupling cap, so that the pot a) doesn't have to run DC and b) doesn't impact the first tubes' anode voltage.

I think I see that you have grounded the RCA inputs at the case? Is this a metal case? Rule#1: ground [is not equal to] ground.

Isolate the RCA's, and use the cable screen to connect their ground lugs to the bottom of the pot, the signal coming from the summing 10k's to top of the pot. Now, the improtant question is: where to ground that pot bottom? The best would be the same (physically the same) point where you ground the grid resistor which should also be the same point where the cathode network is grounded as well as the cable screen from the cable that goes from pot wiper to the grid.

Jan Didden

[gofar99]

Greetings Everyone, I just completed a variation of the amp and it is dead quiet. I'll post the schematics this evening when I get home. Mine is different in that it uses a SRPP driver (ECC802S) and an SMPS for heater power. I have a complete write up nearly ready to go out on the diyaudioprojects web site. My initial thoughts on the one you did are the inputs run too close to the power transformer. I would relocate them to the opposite side of the case, or better yet next to the volume control. I eliminated the input coupling cap on mine as there was nothing that had a dc ofset driving it. One less part in signal chain to pick up hum. As noted by others the input jacks should be isolated from the chasis and the grounds returned to central signal ground. I'll look over your photos carefully this evening and see if I can provide so more insight. My amp works extremely well and baffles many people on how the outputs work.

[Simpleton]

Ok, the enclosure is all made of plastic.
The ground from the RCAs goes to the main star ground, and the shield from the input cable goes to the RCA tab ground. I'm not using a coupling cap for the input of the first grid, it's directly coupled to the source. The first tube's grid is also floating; when i connected a 1M (if memory serves me) resistor to the grid, the hum/buzzing dissapeared but I also lost a lot of volume (nearly 2/3).
The transformer beneath the RCA is the output transformer, the mains transformer is packed beneath the 2 large caps. Both transformers are set at 90º from each other.
Could the PSU bleeder have any influence in this?

Thanks for all the help

[gofar99]

Hi, I'll review your photos carefully this evening (only noon here now) and see what might be amiss. The bleed is unlikely to cause the problem. I ran into a slight power supply filtering problem on the driver stage and had to change out the caps I had there. The case on mine is wood and plastic so by itself that is not necessarily a problem. Are the transformer cases grounded? If so to what? Just a shot in the dark here, but is the amp located near anything that can generate hum. Many months ago I had a preamp that seemed to have a hum problem. It turned out that it was because it was too near my HDTV. The TV was putting out a field when it was turned off. Also, I don't recall if you said anything was attached to the inputs. Depending on what and how things are connected, you can get a ground loop. Back to you later

[Simpleton]

Not near any RF sources, only 70cms away from a lamp dimmer inside the wall which has caused no problems so far.
I have tried the amp at a friends house, again with no RF sources nearby, only the CDP and speakers. Same result.
Thanks

[gofar99]

Hi Again, After looking at the schematic you posted, version one is not good. the volume control shouldn't be in that part of the circuit. It should be relocated to the position as shown in figure 2. The grid on the driver tube needs to be referenced to the ground. The volume control (when up front) will do that. Otherwise the driver will have to get the dc reference from the input device , bad idea. With nothing attached it would most certainly hum. In the proper location the gain would go down, but should be sufficient. I'm using a rather low gain stage and it works fine. I'll check back tonight.

Cheers