Welcome, Guest. Please Login or Register
Decware Audio Forums
03/28/24 at 23:25:22 




Pages: 1
Send Topic Print
ZKIT 1 power supply upgrade questions (Read 7960 times)
chris_m
Verified Member
**




Posts: 34
ZKIT 1 power supply upgrade questions
04/30/10 at 02:36:32
 
Hi,
I've been listening to my completed ZKIT 1 for over a month now and continue to really enjoy it.  I plan to commit it to a steel chassis so I don't have to worry about electrocuting our cats while listenign.  I'm considering some of power supply upgrades and seek some information.

Q1. It looks like the power transformer readily allows me to upgrade to 5U4 tube rectification.  Am I right and would I simply tap off the output tube bias (A) before the 1K dropping resistor instead of after it if I do this?

Q2. SE tube amps typically benefit from choke regulation.  Is there a choke spec I could utilize if I'd like to incorporate this?  Is there a Decware choke that can be purchased for this purpose?

Thanks,
Chris
Back to top
 
 

Torii MkII, AN DAC, home built MLTL with Jordan JX92S
  IP Logged
chris_m
Verified Member
**




Posts: 34
Re: ZKIT 1 power supply upgrade questions
Reply #1 - 05/13/10 at 00:07:31
 
Well, I went ahead with inserting a 5U4G in place of the diode full wave, connecting the yellow/black wire to diode side of the 1K 6W resistor. This should match how it is depicted in the SE84C schematic from the product info page. It works of course but there is a drop in voltage.  There are sonic differences too naturally.

With diode full wave:
voltage point A: 342V
voltage point B: 305V

With 5U4G full wave:
voltage point A: 310V
voltage point B: 279V

The differnce in voltages seems consistant given the specs of a 5U4G vs. a 1N4007 diode.

The question I don't have an answer to is how the 320-350V range suggested for the B+ reading (same as voltage point A) can be that high with the rectifier.  The cathode voltage with the 5U4G is 9.3V which is also below the 9.5-12V recommended range.

So I am left to consider the possibility the ZKIT 1 power transformer is not the same as used with the SE84C unless the schematics and the component values are not current.  I had another 5U4G to try as well.  The voltages were with in a volt of the first rectifier tube.
Back to top
 
 

Torii MkII, AN DAC, home built MLTL with Jordan JX92S
  IP Logged
dank
Seasoned Member
****


pair of dual 18
Imperials

Posts: 418
Re: ZKIT 1 power supply upgrade questions
Reply #2 - 05/14/10 at 13:30:23
 
I have always wondered why the Zkit1 was set so close to the absolute maximum spec of the output tube...now I know!

I'll bet the rectifier tube was replaced by silicon diodes (as a cost savings measure) and no other circuit changes were made.  This resulted in a plate voltage of 346 on the output tubes (342 in your case) instead of 305 in the Zen SE84C.  Being that the cathode is around 10v, this yields a net plate voltage of 336 for the Zkit1 and 295 for the Zen.  In my 6P15P data sheet it lists the maximum permissible plate voltage at 330...always wondered about that.
Obviously the tubes can tolerate more, as my Zkit1 has had further mods to it that boost this voltage even higher, but I suspect tube life suffers, and maybe sound quality changes....OR MAYBE HUM GETS LOUDER!!!!!


Dan  
Back to top
 
 
  IP Logged
chris_m
Verified Member
**




Posts: 34
Re: ZKIT 1 power supply upgrade questions
Reply #3 - 05/14/10 at 21:50:57
 
Dan, that makes a lot of sense to me.  I didn't know the SE84C could have a plate voltage of 305V.  I've never had that amp to examine and the schematic on the SE84 info page lists it higher.  At 305V those tubes should last a good while.

I would guess with higher the B+ goes higher the plate currents and I suspect the hum gets louder.  In fact, the hum filtering probably is tuned closely to the current requirement.  

I am tempted to consider a choke with a little more elaborate RC filtering to raise the B+ at the plate while still limiting hum to decent levels.  But I think Steve's design is highly optimized and it might be difficult to pull that off.

Chris
Back to top
 
 

Torii MkII, AN DAC, home built MLTL with Jordan JX92S
  IP Logged
chris_m
Verified Member
**




Posts: 34
Re: ZKIT 1 power supply upgrade questions
Reply #4 - 05/16/10 at 01:26:40
 
FYI, I continue to listen to the 5U4G in place of the full wave diode pair.  I reported the B+ voltage drops which was expected but does not match documentation.  

FWIW, with the 5U4G I think the backgroud is quieter, darker with diode rectification. With the 6N1P the lower voltage does seem to reduce the speed and dynamics but I put a 6922 in and quite honestly it sounds awefully decent.  Lots of detail.  Very engaging.  The speed seems to be restored with the 6922.

I considering a multi-section RC filter with lower over all resistance to up the B+ a little and see if I can get back to preferring the 6N1P.  

Has anyone else tried the 5U4G with the ZKIT1?  I would be interested to hear your experiences.

Chris
Back to top
 
 

Torii MkII, AN DAC, home built MLTL with Jordan JX92S
  IP Logged
Crazy Bill the Eel Killer
Seasoned Member
****


Tubes Rule !!

Posts: 233
Re: ZKIT 1 power supply upgrade questions
Reply #5 - 05/20/10 at 19:21:24
 
Hello Chris,
I haven't got much time now, but I've played around extensively with my SE84CSEX and here's a few things you could try.

If you're committed to RC filters in the PS, figure out what R value you need and then use a dbl PI filter ( CRCRC) with the total R value needed divided between the two resistors in the PS.

Also, I don't know what cap values are used, but try lowering the first cap value  and then increasing up the line i.e.  15uf,R,40uf,R,80uf. The lower value first cap will decrease the charging currents through the rectifier ( lower noise ), and the increasing values will lower PS ringing and make it more stable.

Plus adding additional RC sections increases the efficiency of the filter far better than increasing either the R or C values.

Only caveat is lowering the first cap value will also lower your B+, so you may have to play with the cap values and the R values to fine tune it. But that's half the fun.

Plus, if you've got the room, you must get as many of the electrolytics out as possible, and put ASC motor runs in instead. HUGE difference.

Have fun and let us know what you decide to do.

Cheers,                Crazy Bill
Back to top
 
 
  IP Logged
chris_m
Verified Member
**




Posts: 34
Re: ZKIT 1 power supply upgrade questions
Reply #6 - 05/21/10 at 13:05:10
 
Hi Bill,
Thanks for your response.  This is very useful information. The multi-section RC filtering is something I have been thinking about but I am considering the reduce R total to increase the B+ for running off of the 5Ru$ vs. the FW diode. The voltage drops from 342V to 310V at the OPT on my kit.

Question: Can you quote the voltages you get at the B+ connection to the OPT and the plate/cathode connections of the signal tube?

I am not at all familiar with ASC motor run caps.  I look forward to researching these.  I will need to plan for a larger chassis if I use them it looks to me.

Thanks again,
Chris
Back to top
 
 

Torii MkII, AN DAC, home built MLTL with Jordan JX92S
  IP Logged
Crazy Bill the Eel Killer
Seasoned Member
****


Tubes Rule !!

Posts: 233
Re: ZKIT 1 power supply upgrade questions
Reply #7 - 05/24/10 at 15:00:55
 
Hello Chris,
sorry it took so long to get back to you, it's been a crazy weekend.

I'm sorry but I can't check the voltages you need, as the amp is in pieces. I'm quessing, and IIRC, its was 300-305 at the plate, and just short of ten volts dropped across the cathode resistor. But those are guesstimates. If I were you, while I'm still breadboarding, I would hit the output tubes at both voltages you are considering, and give a good listen. Let your ears decide.

One thing I can tell you with certainty is that you are not hearing the 6n1p at anywhere near its true potential. Even though they and the 6922 types are touted as replacements except for the increased heater current of the 6n1p, their operating characteristics are so vastly different that one operating point will not be optimal for both tubes. Steve is definately hitting the tube in a far better region for the 6922 types ( around 90 volts @ couple ma, IIRC ). This is definately a very suboptimal operating point for the 6n1p. Try hitting the 6n1p w/ 200 volts @ 10ma ( 200 ohm cathode resistor ), and you'll never look back ( IMHO, very biased as they are my ears doing the listening ,YMMV ).

I'm also not able to give you any suggestions on actual power supply R values to achieve those voltages, as I went to IT coupling so my PS configs are vastly different than what you would need.

One last thought for now. Before you decide on a final config, make sure you listen to the 6922 @ about 6 ma or so, i.e. 100 volts w/ 480 cathode resistors. To my ears a much sweeter sounding point for that tube.

Have fun playing. There's lots of other stuff to try, but you shoulf breadboard the amp w/ stock values first and then listen to get a good feel for the sound of the amp. Then change one thing at a time and listen some more. Eventually you'll find the combo that floats your boat the highest.

Let us know how it turns out. Moving to the tube rectifier was a very good start.

Heres a link for the ASC caps:

http://www.partsconnexion.com/capacitor_film_asc.html

Do a search over on the Hawthorne forum for ASC caps. Theres some great reading from Eddie V.

Cheers,                   Crazy Bill
Back to top
 
 
  IP Logged
Ben
Verified Member
**




Posts: 3
Re: ZKIT 1 power supply upgrade questions
Reply #8 - 05/14/22 at 12:44:13
 
Hi Chris, Dan, Crazy Bill, Steve and anyone else,

I am about to purchase a Zkit1.  I live in Australia so, to avoid rather large shipping charges, I plan on going with all Hammond transformers, which are available with free shipping in Aus.  For the output transformers most likely Hammond 125ESE 15W with (reflected) input impedance adjustable via output tap and speaker impedance selection.  

But it’s the selection of the power transformer that I am most concerned with.  Dan you commented on Zkit1 using a plate voltage (346+/-10V) so “close” to the absolute maximum spec (330V).  I guess you were being graciously generous because 346+/-10V isn't close to 330V, its above it. Of course you added that the 6P15P valves can tolerate voltages above the specified range.  I did read that EL84 valves in guitar amplifiers are sometimes run above recommended plate voltage, but distort they must and hum they do.  

With this in mind I wonder if you would consider what a Melbourne based chap I tried to purchase a Zkit1 PCB from a few years ago wrote when I asked him about his transformer selection.  He bought the Zkit1 PCBs, but ended up building point to point using a valve rectifier.  He said he used a Hammond 370FX power transformer, the output of which is 550V (275-0-275) at 173mA.  For comparison the Zkit1 manual recommends an Edcor with 600V (300-0-300) at 150mA (or 200mA for the 240VAC version).  

He also wrote:

Quote:
“…the power supply voltages at the input tube (of his amplifier) is 290V.  Decware runs them too hot (350V) and burns out the first resistor.  I installed a 25W resister in mine.”


I have read of a few people reporting the first power resister (1K 5W) burnt quickly and was replaced with a higher wattage one.  I plan something similar.  

Leaving that question aside for the moment, it seems to me that using a 275V input transformer would provide voltages that were 8.33% lower than the specified 300V one and that that might not be such a bad thing irregardless of SS or valve rectification.  It might result in better valve longevity OR MAYBE LESS HUM!!!!!

I have never built a valve amplifier before only USB leads and speakers so I would appreciate peoples thoughts and opinions here.

P.S. thanks Crazy Bill for the motor run capacitor recommendation.  I will look into it as I don’t have a chassis yet, so could likely acomodate those beer can size badboys.  As for changing the voltage and amperes at the input tubes, I wouldn’t know how to do that, so if you could point me in the right direction, that would help.

Kind Regards,
Ben
Back to top
 
 
  IP Logged
bobc
Senior Member
***




Posts: 64
Re: ZKIT 1 power supply upgrade questions
Reply #9 - 05/14/22 at 14:52:15
 
Ive built up a couple different versions of the circuit. This is my experience, and what has worked through some trial and error and a little PSU simulations. Add a pound of salt to this....

In my experience, about 300 volts B+ is a sweet spot. This biases the output tubes at about 25ma with a 150ohm cathode resistor.

A Hammond 272FX 300-0-300 is what I typically use, you just have to be sure and connect the primary according to your line voltage. Wrong tap could swing 20volts more. Ive also used the EDCOR transformer for the ZKIT. I could see how with a stock ZKIT the 275-0-275 and the circuit as designed would be a fit as this might compensate for the higher DC from the diodes.

Rectification if you use diodes, maybe try to add a 120ohm 5W resistor after the diodes before the power supply filter / 1st cap. In my experience at the current this draws, this puts B+ in range closer to that of a 5U4. I also added a CL90 NTC to this arrangement to limit inrush. This put B+ at ~302 with the recommended Edcors.

I dont have experience with motor run caps, but the 2x47uf/1K resistor combo in the standard circuit puts out a little too much audible ripple when close to the speaker cone for my taste. The answer is filtering.

Filtering.... if you use diodes, you can support higher value capacitors. 2x100uf in this filter arrangement. Or you could add a choke or another filter stage too.

With a tube rectifier, I've had good luck adding a 10H 156R choke filter stage, and dropping the 1K resistor to 830ohms 1K/5K 5W in parallel. A CLCRC filter. You can even make the first cap 10uf  to support a wider range of rectifiers.

Another PS mod you can do relates to the 3.3uf/10K decoupling network before the preamp tube. I noted in later iterations of the amp this network has been removed and the 47K resistors changes to 56K. Ive done this to my DIY amps. I liked the change.

One more tweak is to change the 1K grid resistors vs 300ohm. The ZKIT is 1K I believe. I found with the right b+ the 300ohm resistors sound better.

Little voltage tweaks for B+ and the plate of input tube makes for a wide range of options. I found getting B+ at ~300 V is the key.
Back to top
 
 
  IP Logged
Ben
Verified Member
**




Posts: 3
Re: ZKIT 1 power supply upgrade questions
Reply #10 - 05/15/22 at 14:33:30
 
Thanks BobC.  

I see now that using a lower voltage power transformer might get the B+ about correct using diode rectification, but then using a choke to reduce hum would decrease it perhaps to too low a level.  So better to use a higher voltage power supply so that techniques that one would want to use such as chokes (and or valve rectifiers) to reduce hum, but also decrease B+ won't leave one with not enough.  Does that sound correct?

I just spied a site:  
http://www.tube-and-valve-electronics.co.uk/product_information/SV83_Electron_Tu...

that listed 300V as the maximum plate voltage for SV83 (6p15p).  Nevertheless, I am inclined to go with your experience re a 300V B+ sweet spot.  With regard to measuring the B+, I know I can measure it with a multimeter connected to speaker ground and the socket for pin 7 for the 6p15p, but I am curious how does B+ relate to point “A” or “B” on the Zkit1 PCB in the “Pre-Flight Check” in the manual?   https://www.decware.com/newsite/ZKITmanual2020.pdf

With regard to diode rectification, you suggested adding a 120ohm 5W resistor after the diodes before the power supply filter / 1st cap.  Would that be placed after the 1K resistor in series with it?

The CL90 NTC definitely sounds like a good idea.  Where would it be placed?  After the diodes before the 1K power resistor?

With regard to a rectifier valve, I have never heard an amplifier with one so I don't know what I'm missing and I note that the well regarded bottlehead amplifiers don’t have them, so I am inclined towards the cheaper option of diode rectification if it can sound great.  Can you tell me what benefits you got when you went with valve rectification?  

With regard to a choke, it sounds like a sensible way of improving hum.  My, now not working, Transcendent Sound SOB has two chokes.  In a diode rectified Zkit1 where would the choke go and approximately what values would you think might be sensible?

With regard to more capacitor filtering, I am inclined not to go overboard, but your suggestion of two 100uF, I presume in place of the 33uF after the diodes and resistors (?), sounds like a good idea.  Perhaps I will go with the motor run version.  

Removing the 3.3uf/10K decoupling network before the preamp tube and changing the 47K resistors changes to 56K would simplify things a bit too.

With regard to the grid resistors, so with the lower the B+ a lower grid resistance from 1K down to 300 Ohm can be used.  Easily done.

Sounds like I have find out what choke to use with diode rectification that would reduce hum and adjust resistance in the power supply values to get/keep B+ at 300V.  Does that sound right?  

Did you use a program to help with power supply calculations?  If so, may I ask which one?

Kind Regards,
Ben
Back to top
 
 
  IP Logged
bobc
Senior Member
***




Posts: 64
Re: ZKIT 1 power supply upgrade questions
Reply #11 - 05/15/22 at 19:02:54
 
You want 300V clean DC.

So with the 600V Edcors or 272FX, the diode based rectification of the ZKIT schematic has a higher B+. One way to achieve this is with the dropping resistor and NTC after the diodes, before first capacitor. The tube has a built in drop.

To clean DC you can increase capacitance with diode rectification beyond the 47uf in the filter stage after the diode (1K resistor). Say 100uf. You can also add additional filtering stages R/C or L/C. The resistor values will need to be adjusted to get to the final B+. Measured with the circuit loaded after the 1K resistor or what feeds the output transformers.

Here is a simulation program:

https://www.duncanamps.com/psud2/

You'll need to make the output tubes and preamp tubes current sources try 50ma and 8ma to get an accurate simulation. You'll have to figure out how to do that.

I'd suggest if you you are building off the PCB you can build it "as is" and add/subtract these mods, so you have a frame of reference. Up to you!

Enjoy!

Back to top
 
 
  IP Logged
bobc
Senior Member
***




Posts: 64
Re: ZKIT 1 power supply upgrade questions
Reply #12 - 05/15/22 at 19:16:56
 
Back to top
 
 
  IP Logged
Pages: 1
Send Topic Print