I wanted to start a thread to discuss power supply options. The power transformer is a fairly large and heavy component in a tube amplifier. Are there other options?
Why aren't switched mode power supplies common in tube amplifiers?
It dawned on me that active power factor correcting computer power supplies first step up the line voltage to ~400v before stepping it down again to 12v, 5v, and 3.3v. There are other voltages but those 3 are the high current outputs. One other output to make note of is the -12v output. Maybe it could be used for bias. The 400v DC source inside the supply would seem ideal for the output stage of a tube amp. The 12v may be ok for some preamp tubes. Would the 5v be enough for lighting 6v heaters?
I'm not sure how isolated a tube's filament is from the rest of the tube. Could a pair of PP output filaments be wired in series for 12v without effecting the function of the tube?
Another cheap switched mode supply is electronic ballasts used in fluorescent lighting. Maybe it could be used as is or modified for tube use. My concern with it is that some may not be 100% isolated from the ac line.
Thoughts?
Why aren't switched mode power supplies common in tube amplifiers?
It dawned on me that active power factor correcting computer power supplies first step up the line voltage to ~400v before stepping it down again to 12v, 5v, and 3.3v. There are other voltages but those 3 are the high current outputs. One other output to make note of is the -12v output. Maybe it could be used for bias. The 400v DC source inside the supply would seem ideal for the output stage of a tube amp. The 12v may be ok for some preamp tubes. Would the 5v be enough for lighting 6v heaters?
I'm not sure how isolated a tube's filament is from the rest of the tube. Could a pair of PP output filaments be wired in series for 12v without effecting the function of the tube?
Another cheap switched mode supply is electronic ballasts used in fluorescent lighting. Maybe it could be used as is or modified for tube use. My concern with it is that some may not be 100% isolated from the ac line.
Thoughts?
The 400V DC rail of PFC type switching supplies is usually on the mains side and should not be used for anything except driving the PFC circuit in this case.
Supplying DIY circuits directly from the mains without an insulation transformer is dangerous and should be avoided under any circumstances.
Rundmaus
Supplying DIY circuits directly from the mains without an insulation transformer is dangerous and should be avoided under any circumstances.
Rundmaus
Besides, 12V anode voltage does not make a preamp, but a quite annoying effects box instead. Also, 5V is way to low for 6.3V heaters, seriously reducing valve life.
Switch-mode supplies can be used for valve circuitry, but they have to be designed for that purpose. Computer stuff will not work.
Rundmaus
Switch-mode supplies can be used for valve circuitry, but they have to be designed for that purpose. Computer stuff will not work.
Rundmaus
I was thinking the +12v could be used as filament power for tubes like the 12AX7. Obviously, if there is 400vdc available, I would not suggest a 12v anode.
I'll have to look at the supply again but I believe the step-up in the pfc stage was a push-pull invertor and isolated. I'm almost sure the negative leg of a 450v cap was on the "pc side" of the circuit.
Power is power and as long as the voltage and current requirement is met (and isolated). I'm sure there's an alternative out there.
Now that you've explained under-volting filaments is bad... "Could a pair of PP output filaments be wired in series for 12v without effecting the function of the tube?" Does the filament have to "float"? If one side or the other of a filament had a positive or negative dc voltage present, would it effect the tube's function? In other words, if two output tubes filaments were wired in series, you would see 0v(gnd) and 6v on one tube and 6 and 12v on the other. How would this effect the amplifier?
I'll have to look at the supply again but I believe the step-up in the pfc stage was a push-pull invertor and isolated. I'm almost sure the negative leg of a 450v cap was on the "pc side" of the circuit.
Power is power and as long as the voltage and current requirement is met (and isolated). I'm sure there's an alternative out there.
Now that you've explained under-volting filaments is bad... "Could a pair of PP output filaments be wired in series for 12v without effecting the function of the tube?" Does the filament have to "float"? If one side or the other of a filament had a positive or negative dc voltage present, would it effect the tube's function? In other words, if two output tubes filaments were wired in series, you would see 0v(gnd) and 6v on one tube and 6 and 12v on the other. How would this effect the amplifier?
Last edited:
You're pretty much stuck with a big iron transformer of some sort,short of a custom SMPS.
One trick I've used in quick-n-dirty prototypes and such is to use a power transformer from a big old SS amp. I had one that was 50-0-50 (0-100),and fed it into a voltage tripler for a B+ of around 320V,IIRC. if the transformer has other windings available,you might be able to use them for filaments/bias/etc.
Really though,whatever you do,the key is ISOLATION.
One trick I've used in quick-n-dirty prototypes and such is to use a power transformer from a big old SS amp. I had one that was 50-0-50 (0-100),and fed it into a voltage tripler for a B+ of around 320V,IIRC. if the transformer has other windings available,you might be able to use them for filaments/bias/etc.
Really though,whatever you do,the key is ISOLATION.
There's a ton of switching supplies out there for anything from laser supplies to lighting. Just have to find it.
I've done the same as you using SS power transformers and voltage multipliers. Mine was from an old Pioneer amp that had a lighting winding perfect for filaments. Then there's the old step-down > step-up +multiplier trick to get filament power.
I understand the importance of isolation and wouldn't suggest otherwise.
That was mentioned. I'm interested in making a small high-power tube amp. Just curiosity.
Don't start giving me ideas for a quad 300b PP amp.
Last edited:
just a question and for others to know...what is the danger using isolated 230vac against direct mains..or rectified 230Vdc? they are of same potential and risky,why need to isolate?
For SS amps its really necessary to isolate and step it down to required voltage.
I see others on the net that some are just using silicon quadrupler circuit to isolate the mains without trafo and some add a series tube rectifier to produce plate voltage., is this Safe?
For SS amps its really necessary to isolate and step it down to required voltage.
I see others on the net that some are just using silicon quadrupler circuit to isolate the mains without trafo and some add a series tube rectifier to produce plate voltage., is this Safe?
Because you don't want to risk any part of the chassis being in contact with the mains. If some part of a homes wiring or even another device's wiring was incorrect and you were to contact them both... It's a bad situation. Better safe then dead. You may also risk damage to other equipment connected to your amp.
I became intrigued with the idea of using a SMPS for the heaters of a power amplifier a couple years ago. Nearly 25% of the power consumed by my Dyna ST-35 is used for the heaters, and since the power transformers in these units run “warmish” anyway, I decided that it might be worth a try. I picked a candidate from my junkbox of computer hardware, and found that it’s easy to modify the 12VDC line from a fixed output to an adjustable output easily capable of 12.6 volts at slightly less than less the 2A required. The 12 volt line on the 220 watt (advertised) SMPS is rated considerably above 2 amps, and with no load on the other outputs, it runs cool, and supplies an amazingly steady, and quiet output. After a couple hours running time, the output settled at 12.62-12.63 VDC and has held that voltage exactly over the 7 month period of use, even if the line voltage is varied from < 105VAC to >130VAC.
Although there’s probably little sonic advantage to running DC on the heaters of a power amp, I feel the power transformer runs cooler, and the “solid” DC may at least be a theoretical advantage for the direct coupled VA/phase inverter. The amount of AC wiring under the chassis is reduced, and in my particular implementation, the heaters are powered on, slightly before the B+ is applied.
Since this worked out well, I’ve modified a couple other supplies for adjustable output; one for a “bench” unit, and another for a tube high level crossover I have underway, and a preamp.
Although there’s probably little sonic advantage to running DC on the heaters of a power amp, I feel the power transformer runs cooler, and the “solid” DC may at least be a theoretical advantage for the direct coupled VA/phase inverter. The amount of AC wiring under the chassis is reduced, and in my particular implementation, the heaters are powered on, slightly before the B+ is applied.
Since this worked out well, I’ve modified a couple other supplies for adjustable output; one for a “bench” unit, and another for a tube high level crossover I have underway, and a preamp.
Yes, but if one side of an isolated 230V ac secondary comes in contact with the chassis, the chassis is at 230V ac relative to the opposite end of the winding. If the hot side of the 230V ac mains contacts the chassis, the chassis is then 230V ac above ground, meaning if you touch it you get nailed because it's hot relative to everything around it.
-Pat
There are quite a few tube amps that use SMPS (one of the sponsor vendor Siliconray actually had one for sale, don't know if he still does or not). Do a search on the forum and you will see a few examples, I built a small guitar amp using an off-the-shelf 12VDC, 5A SMPS, which powers the filaments and a small DC-DC converter (so 12VDC-300VDC). Since the switching frequencies of the SMPS and the converter were both significantly above 20KHz, the amp turned out pretty quiet (the DC filaments also helped). The whole thing costed may be $10...
But if you need higher B+ and higher power, then the choices are tougher, since HV and high-power converters are usually quite expensive and not easily designed or built by the average DIY'er.
Jaz
But not "the same" lethal.
You need to get 230V across your chest to die.
Agree with that.
The **BIG** difference is:
a) transformer isolated amplifier: 230V exists between the chassis (largest thing on the amp) and a few internal points inside: the + end of the rectifier, the positive of the filter caps, the standby switch contacts if available, little else.
If you touch the chassis, or the strings of a guitar connected there, or a microphone metallic frame or screen or any of the speaker wires , NOTHING BAD HAPPENS.
You are touching only one pole , no difference of potential across you, YOU LIVE.
The only way to die is to touch chassis with one hand, with the other touch one of the few points with 230V.
You must be suicidal or stupid to do this.
while in:
b) live chassis.
***EVERYTHING IS 230V HOT***
You touch *anything*, YOU DIE
Specially, you touch the chassis, or the strings of a guitar connected there, or a microphone metallic frame or screen or any of the speaker wires , YOU DIE.
Why?
Because YOU ARE ALREADY TOUCHING THE OTHER POLE WITH YOUR FEET.
I had some unfinished project where I used dual PS, ordinary transformer to power from 120V, and invertor without output chopper to power from 12V. Invertor hick-upped on peaks even though they were lower in consumption than rated, but otherwise it worked just fine.
However, attention to layout and wiring has to be even stricter than for ordinary SS rectifiers. Even SS rectifiers are "audible" in case of some quite common wiring errors that present even in some most expensive amps.
However, attention to layout and wiring has to be even stricter than for ordinary SS rectifiers. Even SS rectifiers are "audible" in case of some quite common wiring errors that present even in some most expensive amps.
This idea still uses a conventional transformer, but it can be any low-voltage, common type, and it is used more efficiently than in a straight peak rectifier configuration:
http://www.diyaudio.com/forums/tubes-valves/230433-flexible-supply-idea.html
It works surprisingly well.
i think this will be the best explanation..thanks and hope this will be clear enough for others.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.