I'm obviously biased, but I've been using the 21st Century Maida Regulator in my 300B amp for over a year now. I'm quite pleased with it. It replaced my older regulator design and the resulting sound quality blew me away.
First off, there's no hum. None. I measure about 200~300 uV RMS (80 kHz bandwidth) of noise/hum/EMI on the output. This with a prototype layout. I expect it to tighten up once I get the amp in a proper enclosure. I've tested the regulator stand-alone with 50 Vpp ripple on the input. The 21st Century Maida reduces this to approx 20 uV RMS noise/residual ripple. If you want ripple rejection, this regulator is where it's at.
Secondly, compared to my previous regulator, I got considerably tighter bass with the 21st Century Maida Regulator. I also noticed that the sound stage seemed to open up and the overall sound quality really benefited from the quiet supply.
I've also tried a regular C-L-C supply. Granted, I didn't go all out on the components - I think I used 470 uF - 2 H - 470 uF, but still... That's 40 dB of ripple attenuation. Still not enough to avoid hum, though. I had about 1 mV of hum on the amp output. I also found the sound kinda fuzzy - perhaps from 60 Hz IMD products.
Shunt regulator in a power amp... Funny.
I wouldn't go there personally as it would require vast amounts of power dissipated in the regulator. For a preamp, sure. But not for a power amp. But then, that's my opinion. Yours is free to vary.I hope this answers some of your questions. The bottom line is that I get the best performance (both subjectively and measured) with a good regulated supply.
~Tom
Good news - especially for international customers.
I am now able to offer more cost effective shipping. The shipping charges are calculated by weight, so cost will vary with the number and size of the boards ordered. The shipping rates for my boards start at:
USA: $5
Canada: $9
Rest of the world: $12
I use quality shipping materials. The envelopes I use are sturdy, well padded, and 100 % recyclable.
~Tom
I am now able to offer more cost effective shipping. The shipping charges are calculated by weight, so cost will vary with the number and size of the boards ordered. The shipping rates for my boards start at:
USA: $5
Canada: $9
Rest of the world: $12
I use quality shipping materials. The envelopes I use are sturdy, well padded, and 100 % recyclable.
~Tom
I built several of these regs last year, but didn't get around to installing in anything until just recently. I put one in my tube phono preamp, replacing a simple mosfet follower on a zener stack... nice improvement, bass is more solid, image is more centered, high treble is much cleaner and less 'splashy'. Running about 35ma at 330v.
I didn't notice when I built it the max capacitance following the reg was 47uF, but I've got 100. No problems though, but I could reduce that to 50 easily, would that be more optimal?
Thanks Tomchr for making this available!
I didn't notice when I built it the max capacitance following the reg was 47uF, but I've got 100. No problems though, but I could reduce that to 50 easily, would that be more optimal?
Thanks Tomchr for making this available!
On the output of the regulator, I'd go as low capacitance as possible. Let the regulator do its job. I usually use a 10 uF electrolytic can where the power goes into the amplifier board. That's the supply bypassing cap and it helps with stability. I'd say 1~10 uF is probably enough supply bypassing.
It's good to know that my regulator can handle 100 uF in your application, though. That's handy for Loftin-White/Ultrapath designs and such.
Thanks for your feedback. I'm glad to hear it turned out well for you.
~Tom
It's good to know that my regulator can handle 100 uF in your application, though. That's handy for Loftin-White/Ultrapath designs and such.
Thanks for your feedback. I'm glad to hear it turned out well for you.
~Tom
If you stay below 500 V, you could use 630 V rated caps. That should cut cost. I recommend using polypropylene caps for the caps on the regulator input and output, C2, and C4. The polypropylene caps aren't cheap ($1~$2.50 as I recall) but they do need to be low ESR and the polypropylene caps fit the bill nicely for that.
Notice the note on the schematic says to populate C1A (expensive polypropylene) for voltages above 400 V and C1B (inexpensive electrolytic) for voltages at or below 400 V.
I can deliver a board with the SMDs populated. I do charge extra for that, though. I'm not clear on which of the SMD parts you want populated. Toss me a PM with the details and we'll work it out.
Thanks,
~Tom
Notice the note on the schematic says to populate C1A (expensive polypropylene) for voltages above 400 V and C1B (inexpensive electrolytic) for voltages at or below 400 V.
I can deliver a board with the SMDs populated. I do charge extra for that, though. I'm not clear on which of the SMD parts you want populated. Toss me a PM with the details and we'll work it out.
Thanks,
~Tom
Hi ska1,
Epcos 1uF 875V polypropylene (Mouser Part No: 871-B32674D8105K) only 3.42 USD (about 2.30 GBP) at mouser.
http://uk.mouser.com/ProductDetail/...hC/EHAdrHxd1Fpg3UVmmlHoPeMuuIorqdE6sboJXfnA==
Epcos 1uF 875V polypropylene (Mouser Part No: 871-B32674D8105K) only 3.42 USD (about 2.30 GBP) at mouser.
http://uk.mouser.com/ProductDetail/...hC/EHAdrHxd1Fpg3UVmmlHoPeMuuIorqdE6sboJXfnA==
Last edited:
Quite a few of the devices in the regulator are in there to allow the regulator to start up into a capacitive load. The requirement of start-up into at least 47 uF load cap also drove the sizing of some of the resistors as significant power is dissipated during start-up and a low-wattage type would not survive.
Some of the devices in the design are there to protect in case of inductive flyback from poor connections - like a cold solder joint on a tube socket.
I had fun torture testing the regulator, that's for sure....
But having an inrush limiter like the CL-80 in the primary circuit of your power transformer will certainly make it easier on the "survival" devices in the regulator. Inrush limiters are good things...
~Tom
Some of the devices in the design are there to protect in case of inductive flyback from poor connections - like a cold solder joint on a tube socket.
I had fun torture testing the regulator, that's for sure....
But having an inrush limiter like the CL-80 in the primary circuit of your power transformer will certainly make it easier on the "survival" devices in the regulator. Inrush limiters are good things...
~Tom
Hi Tom,
I follow the thread and want to experience your combined STW12NK95Z and LT3080 improved regulator for 300B (420v) with friends in FRANCE.
Do you think I need better 2 units for separate lines or can I combined one regulator for two lines? If so then what about caps after regulator?
By the way, I want to use exclusively MKP capacitors;
I want to try the regulator in 2 situation: with valve (274B which needs 4µF max or 5U3C with less internal resistor) and with silicium diodes.
Have you any advice for good mounting? and any PCB still available?
thank you
FRED
I follow the thread and want to experience your combined STW12NK95Z and LT3080 improved regulator for 300B (420v) with friends in FRANCE.
Do you think I need better 2 units for separate lines or can I combined one regulator for two lines? If so then what about caps after regulator?
By the way, I want to use exclusively MKP capacitors;
I want to try the regulator in 2 situation: with valve (274B which needs 4µF max or 5U3C with less internal resistor) and with silicium diodes.
Have you any advice for good mounting? and any PCB still available?
thank you
FRED
Fred,
I have plenty of PCBs in stock. No worries there. Just place an order on my website.
You will need a separate regulator for each supply voltage in your circuit. But as long as the "lines" are the same voltage, you can combine them onto one regulator -- as long as the total current draw is within the capabilities of the regulator. For example, you can use one Maida Regulator to power the two channels in a stereo amp as long as the total current draw is below the maximum for the regulator (see my website for details).
You don't need any additional capacitors after the regulator, though, the regulator will survive if you do have a cap on its output. I have tested up to 47 uF and another builder reported that 100 uF worked as well.
The regulator needs to be fitted with a heat sink. If you're using it in a preamp (low current draw), it may be enough to bolt Q1 to the chassis with an isolating washer (SilPad, mica+goop, or the like) between Q1 and the chassis. If used in a power amp, you'll likely need a real heat sink mounted externally. On my website, you'll find an Excel sheet that can be used to calculate the size of the heat sink for a given current draw, output voltage, etc.
Thanks,
~Tom
I have plenty of PCBs in stock. No worries there. Just place an order on my website.
You will need a separate regulator for each supply voltage in your circuit. But as long as the "lines" are the same voltage, you can combine them onto one regulator -- as long as the total current draw is within the capabilities of the regulator. For example, you can use one Maida Regulator to power the two channels in a stereo amp as long as the total current draw is below the maximum for the regulator (see my website for details).
You don't need any additional capacitors after the regulator, though, the regulator will survive if you do have a cap on its output. I have tested up to 47 uF and another builder reported that 100 uF worked as well.
The regulator needs to be fitted with a heat sink. If you're using it in a preamp (low current draw), it may be enough to bolt Q1 to the chassis with an isolating washer (SilPad, mica+goop, or the like) between Q1 and the chassis. If used in a power amp, you'll likely need a real heat sink mounted externally. On my website, you'll find an Excel sheet that can be used to calculate the size of the heat sink for a given current draw, output voltage, etc.
Thanks,
~Tom
I offer two options:
1) Bare board, no parts.
2) Bare board + STW12NK95Z (used for Q1). I'm told it's hard to get in Europe. Aside from this device, no other components are included.
By CMS parts do you mean the surface mount devices (SMD)? I can supply a board with the SMD parts populated. Toss me a PM for the details.
~Tom
1) Bare board, no parts.
2) Bare board + STW12NK95Z (used for Q1). I'm told it's hard to get in Europe. Aside from this device, no other components are included.
By CMS parts do you mean the surface mount devices (SMD)? I can supply a board with the SMD parts populated. Toss me a PM for the details.
~Tom
I intend to join the "26 tube preamp" DIY palls
Do you have any recommendations using your regulator with this DHT?
Sweet!! With any DHT - and especially when used in a preamp - definitely use a regulator for the B+ and one for the filament. That's the only way I've found to get acceptable hum levels.
I've tested the 21st Century Maida regulator at low currents, high currents, small steps in current at low currents, etc., etc.... That regulator has seen quite a bit of torture testing and passes the tests with flying colors. I have no doubts that it will perform well in a preamp application - including the 26.
My Universal Filament Regulator would be suitable for the filament voltages too...
I'm afraid surface mount components weren't covered in my French class in high school...
I could guess the meaning of the TLA from the context, though.~Tom
Output cap sizing
Hi Tom, Last year I constructed copies of your regs for comparison with my existing designs. Yours are smaller, cheaper, easier to build and work every bit as well as my grossly over-designed behemoths.
What a neat little implementation on the 3080's adj line.. I also conducted fairly rigorous testing with various caps and current draws, these showed your regs to be stable with 200uF output caps @ 420V /150ma loadings.
I would suggest that people not substitute for the STW12NK95Z when using input voltages >480V.
Another caveat is that I always use NTC's on the mains input to the trans to reduce the Pdiss of the FET's during start-up.
How long your thread has become: (hope there wasn't anything too interesting in the intervening pages)!
Thanks for your efforts. Regards, Glens
Hi Tom, Last year I constructed copies of your regs for comparison with my existing designs. Yours are smaller, cheaper, easier to build and work every bit as well as my grossly over-designed behemoths.
What a neat little implementation on the 3080's adj line.. I also conducted fairly rigorous testing with various caps and current draws, these showed your regs to be stable with 200uF output caps @ 420V /150ma loadings.
I would suggest that people not substitute for the STW12NK95Z when using input voltages >480V.
Another caveat is that I always use NTC's on the mains input to the trans to reduce the Pdiss of the FET's during start-up.
How long your thread has become: (hope there wasn't anything too interesting in the intervening pages)!
Thanks for your efforts. Regards, Glens