Remember, it’s always easier to lower the voltage than raise it beyond 1.4x the input AC. Unless you do a voltage multiplier….but…
Ive been using SS rectification for many years now with great results. It has the side benefit of not putting the load of the tube rectifier on the power transformer allowing it to run cooler. Add it is much cheaper than a tube! And you can use larger caps! That a lot of ands…
Ive been using SS rectification for many years now with great results. It has the side benefit of not putting the load of the tube rectifier on the power transformer allowing it to run cooler. Add it is much cheaper than a tube! And you can use larger caps! That a lot of ands…
Whether toobz or SS we need to be a bit careful with that,
Otherwise the RMS rating of the PT or a wdg might be exceeded, 😱
By load of the tube rectifier I meant the 5v winding. The extra 10 to 15 watts not being drawn from the pt if SS rectification is used let’s the pt run easier and cooler.
Generally speaking, when sourcing a power transformer, is it better to opt for lowering heater voltages or lowering the high voltage B+? These two transformers are close, but one has a higher than specified B+ (420V when I'm looking for 360V), and the other transformer has only 4V 3A heaters for the 2A3s which I would need to drop to 2.5V.
Consider:
420V x 1.4 = 588V peak
360V x 1.4 = 504V peak
You only need the filament circuit resistor to drop 1.5 Volts
4V - 2.5V = 1.5V
A 2A3 draws 2.5A
1.5V/2.5A = 0.6 Ohms.
0.6 Ohms x 2.5A squared = 3.75 Watts A 10 Watt resistor is small, and will get hot. Use a 20 Watt 0.6 Ohm resistor.
You can only use the 4V 3A winding to power 1 (one) 2A3 filament.
These are going to be Mono-Block amplifiers?
If you ever decide to use DC filaments for the 2A3, a 4V 3A winding will get extremely Hot!
AC current for a brute force DC supply using a schottky bridge and Capacitor input filter is approximately 2X the DC current of the load.
5A AC current to get 2.5A DC current. The current squared transient integral requires a filament winding current rating to be about 2X the DC current.
For a regulated 2.5V DC filament supply, you will need the filament winding to be much more than 4VAC.
My advice, get a proper transformer, don't attempt to have to do lots of modifications to Shoe-Horn a power transformer into your amplifier.
420V x 1.4 = 588V peak
360V x 1.4 = 504V peak
You only need the filament circuit resistor to drop 1.5 Volts
4V - 2.5V = 1.5V
A 2A3 draws 2.5A
1.5V/2.5A = 0.6 Ohms.
0.6 Ohms x 2.5A squared = 3.75 Watts A 10 Watt resistor is small, and will get hot. Use a 20 Watt 0.6 Ohm resistor.
You can only use the 4V 3A winding to power 1 (one) 2A3 filament.
These are going to be Mono-Block amplifiers?
If you ever decide to use DC filaments for the 2A3, a 4V 3A winding will get extremely Hot!
AC current for a brute force DC supply using a schottky bridge and Capacitor input filter is approximately 2X the DC current of the load.
5A AC current to get 2.5A DC current. The current squared transient integral requires a filament winding current rating to be about 2X the DC current.
For a regulated 2.5V DC filament supply, you will need the filament winding to be much more than 4VAC.
My advice, get a proper transformer, don't attempt to have to do lots of modifications to Shoe-Horn a power transformer into your amplifier.
Can't the voltage after the rectifier be dropped with a resistor before the first cap and choke? Like in this old Asano 2A3... I'd be using a 5AR4 instead of the 83V and feeding it 420V instead of the 380V... Or is this just too much?
That transformer has two 4V 3A filament circuits so it should take care of both 2A3s - as long as I drop the voltage to 2.5V - as I intend to make a stereo amplifier.
I was hoping it wouldn't be a lot of mods... just bringing down the filament or the B+ voltage. Not both... Most of the search is with Hashimoto being reluctant to do a full custom PT... I already need a 120V primary.
Anyhow thanks again for your thoughts!
Another way to reduce an AC voltage, for your scrap book.👍
The 0,05R is the transformer winding resistance. The One R load is equivalent to a 2A3 filament, 2,5A at 2.5V.
The meters are measuring RMS. The diodes simply shorten the conduction period for both positive & negative of the AC waveform.
But lots of harmonics of the line frequency, not suitable for some loads.
I put another scheme on DIY a while ago, I'll pull it up later, 😀
Attachments
Filaments:
I built brute force DC 2A3 filament supplies.
I only had 6.3V windings. So . . .
A Schottky bridge, 22,000uF cap, high-power series voltage dropping resistor, and another 22,000uF cap.
Only 2mV ripple.
Hint: Use a Schottky bridge, it has less voltage drop than conventional silicon diodes, and that allows a higher resistance voltage dropping resistor, which in turn gives more filter action with the second 22,000uF cap that follows the dropping resistor.
B+
Want to reduce B+ voltage?
Move the input cap, 20uF to right After the choke.
Then, put a small cap before the choke (1uF, 2, 3, 4uf? as needed).
Take out that 500 Ohm resistor after the rectifier.
Then put in a 100 Ohm series resistor after the choke, and a 50uF cap after the 100 Ohm added resistor. That will get the ripple voltage back down to as good as, or better than it was in the original circuit.
And . . . your power transformer will run Cooler!
:^)
I built brute force DC 2A3 filament supplies.
I only had 6.3V windings. So . . .
A Schottky bridge, 22,000uF cap, high-power series voltage dropping resistor, and another 22,000uF cap.
Only 2mV ripple.
Hint: Use a Schottky bridge, it has less voltage drop than conventional silicon diodes, and that allows a higher resistance voltage dropping resistor, which in turn gives more filter action with the second 22,000uF cap that follows the dropping resistor.
B+
Want to reduce B+ voltage?
Move the input cap, 20uF to right After the choke.
Then, put a small cap before the choke (1uF, 2, 3, 4uf? as needed).
Take out that 500 Ohm resistor after the rectifier.
Then put in a 100 Ohm series resistor after the choke, and a 50uF cap after the 100 Ohm added resistor. That will get the ripple voltage back down to as good as, or better than it was in the original circuit.
And . . . your power transformer will run Cooler!
:^)
Last edited:
So many ways to skin the proverbial cat.
I'm sure there are some epic threads but is there a short version as to why one might choose DC over AC for filament supplies? I'll probably start with the original schematic and keep it AC and drop from 4V to 2.5V with a series resistor.
For all DHT:
1. AC is easy.
And, with a hum balance pot, you make it as good as you can (but not as good as DC).
2. DC does not have the Intermodulation of 2x power mains frequency sidebands above and below each and every musical tone/note.
DC does not have the Intermodulation of 1x power mains frequency sidebands above and below each and every musical tone/note.
'Nuf Said?
Note: A 3Amp rated filament transformer can not power a 2.5A DC supply.
Sorry.
1. AC is easy.
And, with a hum balance pot, you make it as good as you can (but not as good as DC).
2. DC does not have the Intermodulation of 2x power mains frequency sidebands above and below each and every musical tone/note.
DC does not have the Intermodulation of 1x power mains frequency sidebands above and below each and every musical tone/note.
'Nuf Said?
Note: A 3Amp rated filament transformer can not power a 2.5A DC supply.
Sorry.
I can recommend this very useful software - DuncanAmps PSU-II
If you can accurately enter the data for your mains transformer, you will find that it is quite accurate. Ensure current draw is for TWO channels.
Ditch the 500 ohm resistor in the power supply if at all possible. If you can get away from Cap input supply to a Choke input supply you will also be quite pleasantly surprised.
Don't worry about DC filaments until you get a good design.
If you can accurately enter the data for your mains transformer, you will find that it is quite accurate. Ensure current draw is for TWO channels.
Ditch the 500 ohm resistor in the power supply if at all possible. If you can get away from Cap input supply to a Choke input supply you will also be quite pleasantly surprised.
Don't worry about DC filaments until you get a good design.
Thanks for this! Clear and concise... is there a baseline rating for a 2A3 DC filament supply?
Very cool! downloading the software now.
If you are wanting to do Choke input supply instead of cap input, you might consider the Hashimoto PT-240
One of the problems with buying a mains transformer "blind" with no datasheet is that you will need to do measurements on it and find out what is possible and not. This is one reason that early on I went for Hammond early on. Their specs sheets have all the info you need to simulate your power supply in PSU-II in advance. Example Hammond specs sheet: Hammond 279X
Note the DC winding resistances and off-load voltages that Hammond specs out so nicely. That is very handy when simulating your power supply. And simulating it saves you a LOT of frustration when you go to build.
Later, I sourced power transformers from Thomas Mayer in Germany (but they need a cover). I live fairly close to him so am able to drive there and back in an afternoon or morning.
One more note - don't feel you need to feed the 2a3 filaments from the same power transformer. I don't. This gives you a more options.
One of the problems with buying a mains transformer "blind" with no datasheet is that you will need to do measurements on it and find out what is possible and not. This is one reason that early on I went for Hammond early on. Their specs sheets have all the info you need to simulate your power supply in PSU-II in advance. Example Hammond specs sheet: Hammond 279X
Note the DC winding resistances and off-load voltages that Hammond specs out so nicely. That is very handy when simulating your power supply. And simulating it saves you a LOT of frustration when you go to build.
Later, I sourced power transformers from Thomas Mayer in Germany (but they need a cover). I live fairly close to him so am able to drive there and back in an afternoon or morning.
One more note - don't feel you need to feed the 2a3 filaments from the same power transformer. I don't. This gives you a more options.
Last edited:
6A3sUMMER will probably disagree with me here, but IMHO doing DC filaments is something that you can implement later.
If you hope to buy a power transformer in Japan, note that your chosen primary (120VAC) is not going to be available. Special order in advance is a must. You can save on shipping costs. I would however advise going for a locally sourced power transformer.
While in Japan on business I purchased many Interstage and OPT's. I even found used pairs in excellent condition in Akihabara, Tokyo electronics town. The iso Tango XE-20S is a nice OPT to look out for.
If you hope to buy a power transformer in Japan, note that your chosen primary (120VAC) is not going to be available. Special order in advance is a must. You can save on shipping costs. I would however advise going for a locally sourced power transformer.
While in Japan on business I purchased many Interstage and OPT's. I even found used pairs in excellent condition in Akihabara, Tokyo electronics town. The iso Tango XE-20S is a nice OPT to look out for.
Thanks for the heads up @soulmerchant. I've been emailing a vendor in Akiba that I bought some output transformers from the last time I was there. They'll wind the primary for 120V with enough notice and with the Yen weak, it's a great time to buy.
The PT-240 looks great but goes back to my original question of giving the rectifier (most likely a 5AR4) more AC voltage than the schematic calls for and then needing to reduce the B+...
I am keener and keener on the idea of the Baby Ongaku (see attached PS schematic) as opposed to the original Asano Loftin White... The Baby Ongaku has a choke input which as you mentioned might be a nice match for the PT-240 though it outputs 420V when this design is asking for 360V on each leg.
soulmerchant,
I agree.
joneci,
A 2A3 with AC filaments works very good. But the AC filament related results vary from tube to tube. Select tubes.
A famous quote is: "I never heard a 2A3 amplifier I did not like."
You can always build a DC filament supply Later. You will get even better performance.
You have to have room for one or two electrolytic cap(s); Schottky bridge rectifier; voltage dropping resistor or regulator.
The AC filament winding has to have enough voltage for the Schottky rectifier voltage drop, and the dropping resistor (to isolate the input cap and output cap, and to reduce ripple), or enough voltage for the voltage regulator drop.
The AC filament should be rated for 5 Amps for one 2A3's 2.5 Amps DC.
The DCR of a transformer primary *, and DCRs of all secondaries, and the load on all secondaries, determines the final voltage out.
Does all simulation software take * into account?
Happy building, and happy listening!
I agree.
joneci,
A 2A3 with AC filaments works very good. But the AC filament related results vary from tube to tube. Select tubes.
A famous quote is: "I never heard a 2A3 amplifier I did not like."
You can always build a DC filament supply Later. You will get even better performance.
You have to have room for one or two electrolytic cap(s); Schottky bridge rectifier; voltage dropping resistor or regulator.
The AC filament winding has to have enough voltage for the Schottky rectifier voltage drop, and the dropping resistor (to isolate the input cap and output cap, and to reduce ripple), or enough voltage for the voltage regulator drop.
The AC filament should be rated for 5 Amps for one 2A3's 2.5 Amps DC.
The DCR of a transformer primary *, and DCRs of all secondaries, and the load on all secondaries, determines the final voltage out.
Does all simulation software take * into account?
Happy building, and happy listening!
Last edited:
This is great info and I think it's the play. I'd love to be able to try a DC filament supply in the future so it sounds like the Hashimoto PT-240 is perfect for this with 5.5 Amps per 2.5V circuit.
PT-240
S.480V-420V-240V-0-240V-420V-480V AC170mA
0-2.5V-5V 5.5A 2 circuits
6.3V 3A
5V 3A
Is there anything else I'm missing regarding implementing this transformer in the Rankin Baby Ongaku? Hope to order this weekend... I guess its about 30-60 days to wind it for 120V.
Modern 5AR4s don't seem to be as robust as in days gone by.
Here is a mod that helps protect them from reverse voltage by adding a couple of SS diodes.
From Dynaco ST70 mods
Here is a mod that helps protect them from reverse voltage by adding a couple of SS diodes.
From Dynaco ST70 mods
Pls tell this Geezer of age 90 what is unique about this circuit that the designer feels must be protected.😀
There has been little new in toobz for about 70 yrs. All the valuable patents have expired in most countries.
Except maybe in the US, where anyone can sue for any reason, If Lee Iaccoca of Ford & Chrysler were alive we could ask him.👍
The circuit should perform OK. But will not be outstanding,
There has been little new in toobz for about 70 yrs. All the valuable patents have expired in most countries.
Except maybe in the US, where anyone can sue for any reason, If Lee Iaccoca of Ford & Chrysler were alive we could ask him.👍
The circuit should perform OK. But will not be outstanding,
Attachments
