SEX cathode bypass resistor

[boulos]

Hi,
Would it make a difference to change the Rbias resistor?  If so, would a wirewound resistor be desirable? Does it need to be non-inductive?
Thanks!
Boulos

 

[Grainger49]

Non inductive resistors instead of inductive wire wound resistors are almost always a good thing; but not in power supplies.  The cathode resistor is not in the signal circuit.  The cathode bypass cap is.

 

[boulos]

I'm finding it very difficult to find a non-inductive wirewound resistor at this value and rating. That's why I ask.

 

[Grainger49]

I don't think it is worth worrying about since it is not in the signal path.  The bypass cap is in the signal path.

 

[boulos]

Got it. Thanks!  I want to change the caps, so was going to change the resistor too (it's easier).  So it also doesn't matter if it's wirewound or metal film?

Speaking of the caps. I'm debating between the Simlic II and the Nichicon HE.  I'm sure there are many opinions here, and I've seen the earlier threads. But no one has mentioned the HE series.  They have very nice specs.

 

[mcandmar]

I wouldn't use a wirewound for a cathode resistor unless you need them for large heat dissipation, i would much prefer to use a metal film, i.e. the stock one.  I did replace the two bias resistors for the 1st stage as they were mismatched, went for precision Dale resistors with low thermal drift, but its not really necessary to be honest.  Something on my todo list is to replace those 1.27k resistors with an LED/Resistor string to see can i quantify any differences.

As for the bypass cap, i used Silmic II as i have always liked them, but i haven't tried anything else in the S.E.X. circuit to see what effect they really have.  I have a bag of 1000uf 25v caps with a plan to build a circuit to switch between them to see if i can measure/hear any differences while passing an audio signal through them.  Baggy consists of Elna Silmic II, Tonerex, Cerafine, Nichicon KA, KT, KW, KZ, FG, BP Muse, Panasonic FC, FM, FR, Rubycon LXE, Black Gate, Russian K50-6, Tesla LC, and CapXon aka the crapcons.  ...i've been collecting for a while

 

[boulos]

Thanks!  I will then use the Silmic II (while waiting for your review and keep the metal film resistors.  But I may order new ones because it's easier than removing the old caps off the existing resistors. Are these the right components:  Vishay BC PR03000206800JAC00 ?  Any other ones you recommend?

 

[mcandmar]

Should be fine, also a very good chance those are what came in the kit.  Personally i would just snip the old caps off if you cant unsolder them.

 

[Paul Birkeland]

Would it make a difference to change the Rbias resistor? 

It very likely may, but finding a different resistor of the same value will be tough, and the stock part is a good quality metal film resistor to begin with.

If so, would a wirewound resistor be desirable? Does it need to be non-inductive?

This kind of goes with what's above.  The inductive vs. non-inductive selection isn't so important here, but finding a compact wirewound resistor of the appropriate value isn't going to be easy.

 

[boulos]

Thanks PB!  I'm going to try to maintain using the stock resistors. I ordered a pair of these Vishay resistors PR03000206800JAC00 just in case I mess things up.  During my search I found these RW69V681B12 (http://www.vishay.com/docs/30281/rwmilita.pdf) which are the most compact wirewound resistors with the desired value and rating (~0.5in x ~0.25in).

 

[boulos]

I managed to instal the Silmic II on the existing resistors without much difficulty by clipping the existing caps.  While I had the amp upside down, I changed the 10,000uF 10V cap to a Nichicon HE that I already had (10,000uF 16V). I also added the balanced resistor mod to the filament supply, but instead of 100 Ohm resistors, I used 33 Ohm 1% 3W (Vishay CPF3) that I had at hand. 

Combined with the L-pad I detailed in a previous thread, all hum I had is gone.  The details in the music are much more apparent and note endings linger longer.  I wish I had separated the bias cap mod and the filament supply mod to be able to distinguish which one contributed more to the final change in sound.  But with both of them there, the amp went from sounding good, to sounding great.

 

[boulos]

I figured I should try the headphones without the L-pad (duh!).  Even without it, the hum is gone.  Is it the filament resistor mod, or some joint that I re-flowed in the process of modifying? Regardless, this is the happiest I've been with the amp. 

 

[mcandmar]

For me the heater supply mod made the most difference to hum.  In theory the supply is already referenced to ground via the center tap on the transformer, but in practice centering the supply at the DC side is still required.  I also tried doubling the capacitance of the heater supply to half the ripple which made very little difference so its purely as ground reference issue imo.

 

[Paul Joppa]

For me the heater supply mod made the most difference to hum.  In theory the supply is already referenced to ground via the center tap on the transformer, but in practice centering the supply at the DC side is still required.  I also tried doubling the capacitance of the heater supply to half the ripple which made very little difference so its purely as ground reference issue imo.

Just because I think it's interesting, here's what is (I think!) the technical story. The low-voltage winding doubles as an electrostatic shield for the adjacent high voltage winding, so the high AC voltage is capacitively coupled to the low voltage winding. The LV center tap is grounded to the chassis. But the rectifiers disconnect the LV supply from the winding itself, except for the short interval every cycle when they charge the 10000uF capacitor. Thus the heater can drift away from ground (probably due to heater-cathode leakage current)during the disconnect period of 1/120 second. Then when the diodes conduct momentarily the heater gets yanked back to ground. This fluctuating heater voltage is coupled to the cathode through heater-cathode leakage and capacitive coupling. The cathode is just as sensitive as the grid, so any hum at the cathode is amplified just like an input voltage.

Anyhow, that's my explanation.

 

[boulos]

Thanks Paul!  So adding the resistors prevents the heater from drifting away from ground, thus avoiding the fluctuation and hum?

Here's a picture of the two resistors installed.