It can regulate down to around 2,5 V headroom but taking ripple in consideration I would say you need about 5 V headroom. Then your fine with just a big electrolyt before the regulator.
Staffan
Needing 6,3 V regulated takes 6,3 plus up to 2V for rectifyer loss plus 5 for regulator. Divide that with 1.41 gives you around 9,5 on the loaded tranny. Salas choice of 10 was not a coincident.
Staffan
Staffan
say 4 as minimum
Rod's reg is made as filament (cathode!! for DH tube ) sandwiched between CCS and cap multiplier
it's waste using it for indirectly heated tubes ,with absolutely no benefit
What would be the minimum recommended prefilter? I didn't realize this would cost as much as the rest of the amp. Mouser has hammond choke @$34 and Panasonic 500V 220uF caps at $10 each. JJ caps at parts connexion are $14 for 500uF. Also, If in a seperate enclosure, do i put all this in the psu and umbilical to main box or put prefilter in main box with regultor? I am assuming regulator needs to be in main box.
220uF 220R 220uF I would say as minimum cheap CRC (with solid state bridge). When planning a reg it suffices. If you can make it CLC it tends to sound better. But the choke should be a man at >100mA constant current. 5-10H. If you will put the fiter in another box, use a 10uF MKP or MKT very near the reg's input to ground for decoupling.
Hammond choke at mouser is 20H and 100mA. I looked at the Panasonic ED caps which are 450V 470uF. If this looks good, I will pull the trigger. I tried it all in Psud. Not a lot of difference using more or less capacitance on output, but i cant say i know what I am looking for. Its wierd, the filter with 200uF seems to reach regulated voltage faster than lower capacitance. This is just a result of greater charge time for larger cap, correct. Are you using seperate PSU and how is it arranged? tube rectifier recommendation or will any of the schematic list work equally? I think I read that 5U4G doesnt charge as quickly abd helps with v spikes, or is it current spikes?
With tube rectifiers there is much smaller first cap max spec. Each one notes so in its datasheet. It doesn't work well beyond that max and it expires much sooner.
I assumed a 4.7uF Obbligato Film in Oil in the first cap position since it is what i have, I also simmed the PSU with it. I was speaking of the second cap. I can get BHc aerovox on the Bay, but it will mean waiting. Will probably jst go ahead and get the PANa's.
Manyer high quality tubeamps har been build with SS rectifiers for sure so that will work well for you here as well. While SS rectifiers can be a bit noicier sometimes they certainly have big advantages to. Smaller voltage drop, using the tranny better in Graetz mode for example.
As you have 2 seconday high voltage windings on your tranny I would definately use one for each channel. The diodes are cheap and then you can also do with smaller caps in each filterchannel, say dual 100uF-R-180uF or so. They cost a couple of buck each or so.
Try out something simple like that first, then experiment with snubbers over diodes and some plastics over the electrolyts would be my approach.
Staffan
Last edited:
The diodes I have coming are Fairchild Stealth Diodes, so they are no slackers. I had intended to try dual supplies as you suggested assuming I got the SSHV2 working OK. Couple PSU questions. IS a choke with more cuurent ability better or more Inductance. I can get a 10H choke that can handle 500mA or 20H choke that can handle 100mA. Which is better? Also, while Salas explained why the small cap ws first with a tube, why have a smaller cap first? Is it to reduce inrush?
A choke that is getting near its maximum powerhandling is getting a little less effective in terms of induction. But you seem good within the limits with 100 mA so for two 22(ish) mA lines so the 20H should work for you. If you are going for two filterlines Hammond has good 50 mA chokes to that takes care of your needs.
Though my opinion is that you shall not use more induction then you need for getting rid of PSU noice, so starting with smaller and go up only if you need is my opinion. Sometimes you can live with very small noice and get better dynamics instead.
Smaller startingcaps is used for tube rectifyers simply because bigger is harmful when they backshoot the rectifyer cathode. In SS solutions some like to use smaller 1-2 uF plastic caps as starter as voltage controllers.
Why you usually start small and get bigger in a smoothing filter line is a way of having a "calm and controlled" smothification (nice word) over many steps.
But there are many schools, small caps, big caps, low induction, high induction. Start with an C-L-C or even a C-R-C and listen. Increase or do modifications when you think you need it for a reason, and see if your modifications helps or introduces new problems.
Staffan
PS
If you like tube rectifier but lacks the voltage, a hybrid is possible:
Lundahl Transformers - Hybrid power supply
Staffan
If you like tube rectifier but lacks the voltage, a hybrid is possible:
Lundahl Transformers - Hybrid power supply
Staffan
PSU test setup measurements
I made a test setup with some old output trafos from old radios as second chokes today. I made measurments in circuit of the 6V6 breadboard. The measurements was on a 1 channel C-L-C-L-C-R-C line speced at 16uF-5H-220uF-11H-220uF-2,25k-220uF. There is one filterline per tube.
As we speak of PSU filtering I thought it might be helpful to share.
First, heres the voltages:
Trafo to 5Z3: 2*337 VAC
On first cap: 400 VDC
Second cap: 394 VDC
Third cap: 388 VDC
B+: 336 VDC
And heres the measured ripples:
On first cap/rectifier cathode: 116,1 mV
Second cap: 5,8 mV
Third cap: 5,24 mV
B+: 4,86 mV
What we see here is that the first C-L takes 95 % of the camelbacks away. After that it comes to fine smoothing and small spike eating. This is a decent value I think but it can do with some more work.
Heres a measurement pic on the first cap. scope on AC mode:
And here one on B+:
Staffan
I made a test setup with some old output trafos from old radios as second chokes today. I made measurments in circuit of the 6V6 breadboard. The measurements was on a 1 channel C-L-C-L-C-R-C line speced at 16uF-5H-220uF-11H-220uF-2,25k-220uF. There is one filterline per tube.
As we speak of PSU filtering I thought it might be helpful to share.
First, heres the voltages:
Trafo to 5Z3: 2*337 VAC
On first cap: 400 VDC
Second cap: 394 VDC
Third cap: 388 VDC
B+: 336 VDC
And heres the measured ripples:
On first cap/rectifier cathode: 116,1 mV
Second cap: 5,8 mV
Third cap: 5,24 mV
B+: 4,86 mV
What we see here is that the first C-L takes 95 % of the camelbacks away. After that it comes to fine smoothing and small spike eating. This is a decent value I think but it can do with some more work.
Heres a measurement pic on the first cap. scope on AC mode:
And here one on B+:
Staffan
Last edited:
The hybrid rectifier comes handy when the HT winding of the power transformer don't have a center tap. However I don't see how it adds HT voltage.
I might be wrong but Buzzforb has two secondary 300 V windings, ie he could serial them and make the winding connection to center tap. But if he thinks two times 300 gives to low output voltage for his filters and B+ he could instead parallell them and make a hybrid graetz and get a higher output.
Staffan
Thanks for the info stajo. I got the stuff for seperate filters for each B+ using CRC of 4.7uF/220r/470uF. The trafo is a 200Va toroid with dual 300V and 6.3V secondaries. I'll probably use another small trafo for dual filament supply and leave the 6.3's unused until the tube rectifier gets here. If I like the amp, and I most likely will, I will try the tube and choke way as well, just to hear the difference. This will allow me to spread my purchases so the wife doesn't shut the whole operation down. Went with Mills and PRP for resistors, and will use Obbligato FIlm in Oil 2.2uF on output and Silmics as bypass. I'll post pics as soon as I get em. Squeaky wheel one last time, Is seperate PSU better?
I now know why ZM would not sell me Pumpkin pre. He knew what a PIA I was and didn't want to teach me electronics😀
I now know why ZM would not sell me Pumpkin pre. He knew what a PIA I was and didn't want to teach me electronics😀