I have fired up my SSE and there are a severe hum in the speakers, I still hear it when I leave the room with the door open.
This is my first build and the setup are:
The amp are now configurated in triode without feedback (see pict)
The board are grounded via the PCB together with the zero point of T1, the center of 375-0-375.
Incoming signal are grounded via the PCB at the IN-connectors (pin 2 & 4)
Speakers are grounded via the PCB at connector T2&T3 sec
I have not grounded the power and OPT shields, it is on my ASAP to-do list.
Measurments:
Things I consider to try out, in specific order
Will negative feedback help?
Shifting the sec of OPT?
I’m considering a choke. However are this related to cost (I have already exceeded my budget) and lead time, if I order today will I not receive it until next week, the weekend will be spoiled. Therefore is this not my primary solution, more like a backup.
Anything else, what have I missed?
This is my first build and the setup are:
- Power: Hammond 374BX
- JJ 5AR4
- JJ 12AT7
- Russian 6L6GC equivalent (see pict)
- OPT: Edcor GXSE15-8-5
- No choke, R1 = 150 Ohm
- Speakers: No name 8” + 1” tweeters from the 1995-isch.
The amp are now configurated in triode without feedback (see pict)
The board are grounded via the PCB together with the zero point of T1, the center of 375-0-375.
Incoming signal are grounded via the PCB at the IN-connectors (pin 2 & 4)
Speakers are grounded via the PCB at connector T2&T3 sec
I have not grounded the power and OPT shields, it is on my ASAP to-do list.
Measurments:
- B+ = 454 V
- R17/R27 = 680 Ohm, wired to the outside of the PCB for easy exchange during the test period
- U_R17/R27 = 35 V
- I = 35/680 = 52 mA
- P = (454-35)*0,052 = 21,5 W
- VAC in idle on the OPT-sec are 24 mV on left channel and 54 mV on right channel, measured with my DVM.
Things I consider to try out, in specific order
- Ground the casing of transformers
- Twist power wires (5, 6,3 and 375 V)
- Add a capacitor at T1 red/yel-GND
Will negative feedback help?
Shifting the sec of OPT?
I’m considering a choke. However are this related to cost (I have already exceeded my budget) and lead time, if I order today will I not receive it until next week, the weekend will be spoiled. Therefore is this not my primary solution, more like a backup.
Anything else, what have I missed?
Attachments
That layout is on a wood base?
So there is no shielding plate under the PC board?
And no shielded wiring of the input wiring?
So of course it will pick up hum.
Put a sheet of aluminum under the PC board and ground it to the PS ground.
So there is no shielding plate under the PC board?
And no shielded wiring of the input wiring?
So of course it will pick up hum.
Put a sheet of aluminum under the PC board and ground it to the PS ground.
Attachments
Twisted the wires.
Didn't have any extra 500V caps, have to order that. 100-300uF?
Tried to shift OPT
Tried feedback, both negativ and positive to be shure.
NO progress
The humm are initially 100mVAC, goes down to 20-30 mVAC after few minutes.
The frequency are 100Hz (grid are 50). Therefore do I conclude that the origin are in the area of the rectifier or insufficient capacitor.
Yes, I also tried to remove the rectifier and run on diodes. No change.
Didn't have any extra 500V caps, have to order that. 100-300uF?
Tried to shift OPT
Tried feedback, both negativ and positive to be shure.
NO progress
The humm are initially 100mVAC, goes down to 20-30 mVAC after few minutes.
The frequency are 100Hz (grid are 50). Therefore do I conclude that the origin are in the area of the rectifier or insufficient capacitor.
Yes, I also tried to remove the rectifier and run on diodes. No change.
This can't be the first SSE build on wood.
Please explain how a ground sheet below the board would improve the situation?
Ordered recommended or eqvivalent from Digi-key. Any perticular capacitor in mind, i can look up the specific item number if you tell me the position on the PCB.
Attachments
One wavelength are 10ms => 100hz.
Sending an 100Hz signal to input matches the hum. 50hz are significant lower.
