Hoffman Amplifiers Tube Amplifier Forum

Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: Leevi on February 24, 2011, 03:53:36 pm

Title: Hum again
Post by: Leevi on February 24, 2011, 03:53:36 pm
I got an SE bass amp with 6L6 wired and unfortunately there is a low hum
in the background regardless if the volume is open or not. The chassis is quite
small and the big Hammond 1627SEA OT is dominating the chassis.
I'm wondering if I should rotate the lying PT 90 degrees (OT is upright). Do you think
that would decrease/remove the hum.

What is a good way to ground an SE amp:
- everything in one star/bus
- preamp in own bus/star and power amp in own star

Schematic:

/Leevi
Title: Re: Hum again
Post by: VMS on February 24, 2011, 04:18:27 pm
Is the hum 50Hz or 100Hz?

How much are you filtering the plate supply?
Title: Re: Hum again
Post by: phsyconoodler on February 24, 2011, 05:41:12 pm
The schematic is just a picture of the amp head,no schematic.
  I find the biggest factor causing hum in an SE amp is two things:

1) the preamp and power filter caps need to have separate grounding points.

2) the filaments are noisy and need to be well away from other wires,sometimes requires elevated heaters.

Also if it's cathode biased,use the cathode as a ground point for heater center taps or the same for two 100 ohm resistors.

  I follow the layout for Fender's 5F2A and have good results.
Also too hot bias will cause hum.
Title: Re: Hum again
Post by: jeff on February 24, 2011, 08:21:15 pm
I'm wondering if I should rotate the lying PT 90 degrees (OT is upright). Do you think
that would decrease/remove the hum.
Maybe not. What you don't want is the Power transformer and Output transformer in the same orientanion.
if the PT is lying and the OT is upright I think you're good. I don't think rotating will help much. If they were both upright rotating would help more. Is there space between them or are they right next to each other?

 How loud is it? Does it get louder as you turn it up?

 Filtering may be a problem. SEs need more filtering than PPs because in a PP the transformer cancels the hum. You could try this. Disconnect the rectifier from the first filter cap. Add another filter cap from the rect to ground. Add a 250 ohm 5W(don't know the voltages or current draw on your amp so make sure 5W is enough) resistor from one filter cap to the other. It'll drop the voltage a bit but I did this to one of my SE amps and it's dead quite.



----->|--\              (add)         B+
     (rect)  \________/\/\/\/\_______
              /        |                   |
----->|--/        =(add)            =(original)
                       |                   |
                   
Title: Re: Hum again
Post by: Leevi on February 24, 2011, 11:54:50 pm
Quote
The schematic is just a picture of the amp head,no schematic.
Sorry for that, I added the schematic to the first message
Quote
Is the hum 50Hz or 100Hz?
100Hz
Quote
if the PT is lying and the OT is upright I think you're good.
Is there space between them or are they right next to each other?
OK, there is about 8cm between them
Quote
How loud is it? Does it get louder as you turn it up?
It doesn't not increase with volume. Maybe I will next try to add more filtering
Quote
I follow the layout for Fender's 5F2A and have good results.
Also too hot bias will cause hum.
I'll check that

The filtering is already now pretty high. I have also tried to use choke which didn't help at all.

/Leevi
Title: Re: Hum again
Post by: HotBluePlates on February 25, 2011, 03:47:44 am
An output tube screen is somewhat sensitive to hum in its power supply; fortunately the gain from the screen to plate is fairly small.

Since you're using a UL output stage, what about trying an extra stage of filtering (choke or smallish resistor plus a cap), rather than just adding more capacitance to the existing filter stage. Many have had problems with hum in a SE design when using speakers with decent bass repsonse, and adding a stage of filtering often helped.

Switching from UL to normal pentode operation may make adding a filter stage easier. At least the choke could be smaller if it only had to pass screen current. In this scenario, you'd want the filter stage for the plate, then a filter stage, then another filter stage to feed the screen. Sounds like a problem, but the plate doesn't need to be very clean (the signal voltage at the plate will be more than 100v, while ripple will be likely under 10v). The screen node should be quite clean.
Title: Re: Hum again
Post by: VMS on February 25, 2011, 04:51:03 am
I also suggest the use of Pi filter before the plates, especially with UL-amp.

Merlin has a good page about filtering:

http://www.freewebs.com/valvewizard1/smoothing.html (http://www.freewebs.com/valvewizard1/smoothing.html)





Title: Re: Hum again
Post by: LooseChange on February 25, 2011, 05:45:33 am
All of my SE amps have a PI filter utilizing a choke feeding the plate supply.  This was done after finding that in Triode mode the amp hummed.
Title: Re: Hum again
Post by: Leevi on February 25, 2011, 11:56:12 am
Thanks a lot for the advices guys. That helped and I succeeded to get rid of the hum.
I added an extra filer stage and tried both choke and resistors. I ended up to 2 x 100 resistors
instead of the choke since I wanted to drop the voltage. I also removed the NFB in order to get
the sound more fat. The biasing is still little bit open. I'm using 750 Ohm resistor on the cathode
which means 53mA current. Is that OK? See the updated schematic.

/Leevi
Title: Re: Hum again
Post by: phsyconoodler on February 25, 2011, 12:18:14 pm
Are you using 680uf filters? That is a HUGE amount of filtering.I would suggest separating the filter cap ground points.The series caps get grounded near the transformer and the preamp filters ground near the tonestack area.
  With a 345v secondary you will likely have about 480v on the plates,so you need to bias about 90% or so dissipation with cathode bias.
I use the Weber Bias calculator and find out how many watts your tube is dissipating.
   By choosing a very large cathode resistor it depends on waht PT you have and whether the plate voltage sags a lot when you get that power tube cooking.
  e.g. I have a push-pull amp that I wanted to cathode bias but as I increased the cathode resistor size to get it where it wanted to be,the plate voltage was dragged down a LOT.The PT is rated at 200ma so for a pair of 6L6's in fixed bias it works ok,but definitely sags when you try and get some low end out of it.With cathode bias it sagged so much the plate voltage went from 425v to 340v.
  At 35ma fixed bias it works fine as long as you don't dime it.If you do,it sags like a mother.
At reasonable volumes with pedals it really shines.
   Your's is a bass amp,so stiff power supply is definitely the order of the day.But because you want to be able use different power tubes it becomes a compromise.6L6s might run too hot or on the edge and a 6550 will be loafing.
  If it were me I'd make it fixed bias and have a large range of bias voltage adjustment with external ports.Then you could get the most out of each tube used.
Title: Re: Hum again
Post by: Leevi on February 25, 2011, 01:04:43 pm
Phsyconoodler,
Quote
Are you using 680uf filters?
Yes, I'm. Note that the total capacitance is 340uF because the caps are in series.
Plate voltage is 465V.
I used the Weber calculator and the biasing seems to be little bit hot for 6L6 but
I'll keep it there and as you stated
Quote
6L6s might run too hot or on the edge and a 6550 will be loafing.

The sound of the amp is great. The sound is clean even with full volume.
The bypassing of the EQ works as an overdrive.

/Leevi