Kuei Yang Wang said:
i was thinking of taking it from 300b cathode initially.
i noticed in one of your circuits for bias/floating of heater... in order to float the heater of the driver, i'll have to take the resistor divider from 300b to the CT winding of the heater, and then ground the center point with 2 resistors where the c3g filaments are. would this be about right?
Kuei Yang Wang said:
ok. help required here.
now i'm confused as to how to bias/float my c3g heaters.
i initially intended to take the voltage across R1 as shown below(numbers in blue were my design values, in red are measured values.):
An externally hosted image should be here but it was not working when we last tested it.
if i run one resistor from the 300b cathode to heater CT, this would result in a resistor having to drop 244vdc(264v-20v). any suggested values for the R series C from heater CT to ground?
thanks.
Konnichiwa,
Why not use 220K from the 300B Cathode to to C3g Heater centertap, this will add around 1mA current draw to the 300B and should be harmless, a Resistor of 1W Rating is suggested. Then use a 18K...22k (0.5W suffices) resistor to ground from the C3g Centertap.
The minimum capaictance in the heater bias divider would then be a few uF, but why not procure 100uF/50V and just use that?
There is no major issue with some tolerance in that.
BTW, are you absolutely SURE that your noise does not come from the HT line? The DRD circuit originating with jack Elliano has very poor/non at all power supply noise rejection.
You could connect (temporary) a large value electrolytic capacitor bank of suitable voltage rating across the midde capacitor in your supply (55uF) to see if that kills the hum. If so you know it is insufficient +B filtering.
Sayonara
garbage said:
Why not use 220K from the 300B Cathode to to C3g Heater centertap, this will add around 1mA current draw to the 300B and should be harmless, a Resistor of 1W Rating is suggested. Then use a 18K...22k (0.5W suffices) resistor to ground from the C3g Centertap.
The minimum capaictance in the heater bias divider would then be a few uF, but why not procure 100uF/50V and just use that?
There is no major issue with some tolerance in that.
BTW, are you absolutely SURE that your noise does not come from the HT line? The DRD circuit originating with jack Elliano has very poor/non at all power supply noise rejection.
You could connect (temporary) a large value electrolytic capacitor bank of suitable voltage rating across the midde capacitor in your supply (55uF) to see if that kills the hum. If so you know it is insufficient +B filtering.
Sayonara
Thanks for the reply.
so the voltage divider is actually this 220k resistor and a 18-22k resistor to ground, with the c3g heater CT between the 220k and the 18-22k resistor. now it's much clearer. with the voltage divider starting at the top of R1, i should get about 20 - 26vdc,
i have another spare 55uF cap. although not a large value, but i will try putting that at my middle cap to see if it helps.
Kuei Yang Wang said:
so the voltage divider is actually this 220k resistor and a 18-22k resistor to ground, with the c3g heater CT between the 220k and the 18-22k resistor. now it's much clearer. with the voltage divider starting at the top of R1, i should get about 20 - 26vdc,
actually i am not 100% sure that it is not due to sufficient psu filtering. i did try removing the ultrapath cap altogether(without cathode cap at 300b, much much less gain though), but that did not produce much difference in hum/hash levels.Kuei Yang Wang said:
i have another spare 55uF cap. although not a large value, but i will try putting that at my middle cap to see if it helps.
put in an additional 55uF at 2nd cap but the hum remains.
used a voltage divider of 220k and 8k from B+ to CT of c3g(raised by 22vdc). (easier to solder in the parts from B+) hum still there.
running out of ideas now. i am wondering if the psu chokes and power trans orientation is not correct. below are some pics of the chokes and power trans. can some one comment if their orientation looks ok?
An externally hosted image should be here but it was not working when we last tested it.
used a voltage divider of 220k and 8k from B+ to CT of c3g(raised by 22vdc). (easier to solder in the parts from B+) hum still there.
An externally hosted image should be here but it was not working when we last tested it.
running out of ideas now. i am wondering if the psu chokes and power trans orientation is not correct. below are some pics of the chokes and power trans. can some one comment if their orientation looks ok?
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
analog_sa said:
i am using a double pole triple throw switch(got the switchs here) to turn on my circuit.
first, filaments only are turned on (83, c3g, 300b), then i wait till the mercury is heated, before turning on B+.
when filaments are on, there is always complete silence. the hum/hash is only there when the B+ is turned on.
not sure if this means that it is B+ related. if so, is it due to insufficient filtering or emf/rfi interactions between the windings and cores of the power trans and the psu chokes?
There is no way to know if you don't conduct some experiments. What i proposed is very simple. Get the amp nice and warm, make sure that the hum is clearly audible and remove the B+ transformer connection. The amp will continue working for some time thanks to the PS caps but: there will be no B+ ripple at all and there will be no magnetic coupling between transformers/chokes and sensitive nodes in your amp. If this experiment cures the hum (which i can almost bet it will) you'll then have to find the exact way the hum gets in.
analog_sa said:
i see. i'll do just that.
btw, i put in the rectifier, choke(2mH), 4700uF,4700uF,4700uF,4700uF,4.7uF,1R, for the 300b filaments. hum seems lesser.
a measurement via multimeter in ac setting shows 0.011v at the speaker terminals with volume at zero. previously it always showed 0.014v. but the hum is still audible from my normal sitting position.
maybe the hum is from the mur860 rectifiers? will swap in a normal bridge rectifier to see if it helps...
Kuei Yang Wang said:
hi KYW
i am measuring 0.125vac on my 300b heaters.
i tried using a pot at the heaters with the cathode resistor wired to the CT of the pot, but the result was the same as before. i've read that many people tried using pot with great success... not for my case though.
now i have 0.01vac on the speaker outputs. this is with rectifier, choke(2mH), 4700uF,4700uF,4700uF,4700uF,3300uF, choke(2mH), 1R6. interesting to note that before the resistor, it is 0.184vac, after the resistor it is 0.125vac. the dropping resistor was 1R2 (2R2//2R7, 10W each) all along(1R stated in the above thread was my mistake).
i have to agree with analog_sa that changing to bridge rectifiers did not make any difference.
you never know till you try huh?guess i will stick to this configuration for now. with my 93db speakers, i barely notice the hum when the cd is not playing, while seated in my listening position.
thanks all for the help.
kmtang said:
i tried to put my multimeter on ac setting with probes on the B+, but could not get a stable reading.
also, when putting back my 300b heater dc circuit, i added a 10000uF cap to the filtering.
so now, it is 4Xmur860, 2.2mH, 4700uF, 4700uF, 4700uF, 4700uF, 10000uF, 2.2mH, 1R2.
the hum when measured across the speaker output terminals with my multimeter in ac setting, gives 0.007V. i guess most of the hum is introduced by the 300b heater circuit. with insufficient filtering (4Xmur860, 4700uF,4700uF,4700uF,1R2) initially, i get 0.014V.
i can safely say that the hum is not because of the 83.
current 300b heater circuit:
An externally hosted image should be here but it was not working when we last tested it.
Yo garbage:
I notice earlier you tryed a LM338 to regulate the 5v...those required a 3 volt drop to regulate proberly (ie. 8 volts in..5volts out) Did you have the required drop? I built a couple of parrallel 300B SE's and did the following:..use the 6.3v tap-> bridge rect. -> 68,000uF cap(yo!) -> 6.3v filaments and LT1084 (low drop 5A reg) -> 300B filaments. I think I ended up with like 7.5v after the bridge.....the LT1084 only needs 1V drop to regulate. I think the speakers make more noise without the amp connected than with ..
I notice earlier you tryed a LM338 to regulate the 5v...those required a 3 volt drop to regulate proberly (ie. 8 volts in..5volts out) Did you have the required drop? I built a couple of parrallel 300B SE's and did the following:..use the 6.3v tap-> bridge rect. -> 68,000uF cap(yo!) -> 6.3v filaments and LT1084 (low drop 5A reg) -> 300B filaments. I think I ended up with like 7.5v after the bridge.....the LT1084 only needs 1V drop to regulate. I think the speakers make more noise without the amp connected than with ..
MelB said:
I notice earlier you tryed a LM338 to regulate the 5v...those required a 3 volt drop to regulate proberly (ie. 8 volts in..5volts out) Did you have the required drop?
...68,000uF cap(yo!)
...I think the speakers make more noise without the amp connected than with ..
hi MelB
i had a 0-7vac winding for my 300b heaters. there's enough voltage headroom for regulation.
that's a big cap for filaments! looks like i need to look into using bigger caps for the 300b filament supply.
thanks for sharing your experience.