A WE-91B project

[jerrylong]

I have used a roll of copper foil to shield 310A
This is without shield

Hum voltage is 1.57V

And used copper foil to shield V1

Hum voltage drops to 1.238V

Try this way in V2

Hum voltage drops to 0.958V（about 60% of without shield）

 

[jerrylong]

Ver 2.0 PCB arrived today, Begin to assemble it tonight

 

[jerrylong]

Second attempt. Power PCB

 

[jerrylong]

Amp PCB

 

[jerrylong]

Amp PCB nearly finished

 

[jerrylong]

Begin wiring
Power plug to Power trans


Power trans to 274B(HV and heater)


Heater(310A and 300B), UV meter power wire


Use copper foil to shield


Input and output wiring


Put PCBs in

 

[jerrylong]

Begin to test
I have prepared two sets of GND wire to test which GND plan is better


GND plan 1# wire


GND plan 2# wire


testing GND plan 1#


voltage of noise is about 140mV(much better than PCB Ver 1.0)


and I also tested GND plan 2#

noise voltage is quite the same.

 

[jerrylong]

I have tried the following things to reduce the noise:
1, used copper foil to shield two 310A
2, tried two kind of GND method
3, used a 1K VR paralleled to the cathode of 300B and connected the CT to here



noise reduced a lot compare with the previous one, but still too much, is there anyway I can try?

 

[jerrylong]

Testing stage from stage
Starting from 274B output


Vpp=40V(I think it is normal, because has not went through filter circuit)

 

[jerrylong]

Testing the point after the choke


Vpp=13V(I think it is too high?)

 

[jerrylong]

Changed GND point, now the noise has reduced to 17mV

 

[jerrylong]

Used a roll of copper foil as shield, Noise reduced to 8mV, finally succeed

 

[jazbo8]

Congrats! Now you can fire up the amplifier and enjoy some tunes on your TAD clone.

 

[jerrylong]

customized shield received to day, need some Grinding and polishing work

 

[jerrylong]

and one more problem, B+ is now 420V, I need to reduce it to 400V, so I need to add a res(180 ohm) by the way add a 50uf cap to get a CRCLC net to reduce the ripple of the B+, after that the whole project will be finished.

 

[Troy McClure]

I like you customized shields. They look great and are really unique!

By using a 1k VR you have changed the cathode resistance by adding ~500 Ohms. That's the reason why your B+ has increased by 20V because the current through the 300B is much lower now.
You should use a 50 or 100 Ohm VR which can handle some current (>= 1W)
I also wouldn't add a 50uF to the PSU. Stick to the original one since it's influencing the sound of the amp.
In the end your biggest issue will be AC filament supply. If you want to reduce your THD further you have to switch to DC.

And you should measure your frequency response...

 

[jerrylong]

Thanks Troy
1, I haven't added 1K VR to the 300B cathode now, B+ is 420V, so I thought of adding a 180 ohm res to reducing the voltage.
2, I will test FR next week, after I put shield on.

 

[Troy McClure]

1. That will change the characteristics of the 300B and it won't work with proper values (different sound).
You can use the 1k pot temporarily but you should really use 50 to 100 Ohms afterwards...

 

[jerrylong]

Thanks I realized the hum mainly comes from AC of 300B heater, adding a VR（50~100R）is the only solution, my power trans have 5V for the 300B heater and 10V for two 310A, it is hard to switch to DC for 300B

 

[Troy McClure]

If you can live with this AC hum it's fine. It depends on the sensitivity of your speakers at 50/100Hz and how loud it is on your listening position.
I have 3 different 91A clones, all with different filament supplies:
1. AC (5V)
2. simple DC (5.3V AC - bridge rectifier - 10000uF cap - 5V DC)
3. Tentlabs filament supply (6.3V AC input - 5V DC)
AC sounds best (if you ignore hum), second is Tentlabs, last is simple DC.