Rev C troubleshooting

[Thermion]

One problem down, one to go.
:crazy:
When I first hooked up the Rev C I had a bad hum problem. Swapping patch cords helped some, but not completely. The first thing to think is a grounding problem somewhere. In my previous projects, I always used the hot and neutral wires without connecting the earth ground to the chassis. The circuit ground would be tied to the chassis which was allowed to float referenced to ground. In my Rev C project, I connected the PCB ground plane to the chassis, and tied the chassis to earth ground.

After discussing the topic with Shannon, I disconnected the chassis from the PCB ground plane and allowed the circuit to float. Both my input and speaker connections are insulated from the chassis, so this wasn't a problem. The earth ground is still connected to the chassis for safety, and the hum has gone away. I have to hold my head next to the speaker to hear it now.
:toothy8:
The next problem is related to the tube rectifier. I designed the amp to use just about any of the popular rectifiers, but with a 5AR4 in mind. I have high mains voltage here, (125-127 Vac, 60 Hz) and this makes the filament voltages on most tube amps run real hot. I normally have to use a variac to run my gear.

Anyway, in the Rev C amp I didn't want to have to use a variac and Shannon's design takes this into account, except for the filament on the rectifier. This voltage is running around 5.8 volts. Should I just add a resistor in series to lower this, or is there a better way? I am afraid that I will shorten the lives of the rectifiers at these voltage levels.

JT

[Shannon Parks]

Anyway, in the Rev C amp I didn't want to have to use a variac and Shannon's design takes this into account, except for the filament on the rectifier. This voltage is running around 5.8 volts. Should I just add a resistor in series to lower this, or is there a better way?

I think I have an elegant solution for you. Hopefully you have some of the Keystone CL-50s - if not I'll get some to you. These are the 'black capacitor thingies' I usually give away as a freebie in my kits. Checkout:
http://www.diytube.com/cl.pdf

Looking at the datasheet, the part is rated for 1.5A to 5A. I believe you can even derate these and no have problems (eg using instead of CL90 in series with the primary) but it is within current specs for this application. Unfortunately, we are at the lower end - an unlinear, semi-uncharted end. 1.9A divided by 5A tells us that the load is 38% of the max load rating. Look at the chart - at 25% loading it is near 1 ohm resistance and at 50% .4 ohms. I would swag that due to this quick slope from 25% to 50% would mean that 38% would leave you at close to .5 ohms (.5X1.9A = .94V voltage drop). I think no matter what you will get 4.5V or higher, which is +/-10%. Give it a shot and let us know how it goes!

Shannon

[EWBrown]

I'd go with a resistor in series with each leg of the 5V AC to the rectifier tube filament (perhaps the CL-50s might be useful here?) Typically a 5V rectifier tube draws either 2A (5Y3) or 3A (5AR4/GZ34) so figure the resistance to drop the extra 8/10 VAC from there.

After eliminating any ground loops, any further "humbusting" could be accomplished by using one of the 1.5 Henry "C354" replacement chokes from Uncle Ned's, in place of the 50 ohm or (two 100 ohm WW resistors in parallel) , this will have about 64 ohms DC resistance, and will give about 1130 ohms reactance to any 120 Hz ripple on the FW rectified DC voltage.
(XL = 2*Pi*F*L, 6.28 X 120 X 1.5 = 1130 Ohms equivalent)

Also, make sure that the power and OPT transformer cores are orthoganally oriented (90 degrees) so that any AC magnetic fields from the PT don't couple over to the OPTs.

The DIY35s should be totally hum-free, I've built a total of 5 now, (3 Bs and 2Cs) and none have any hum at all. The original Dynaco ST-35 has a tiny amount of hum, but then the filter capacitor is old, and of a much lower capacitance.

/ed brown in NH