I could use some help with specs for a set of power transformers for my DCPP mono tube amps. I am struggling to get this spec'd right
What the amplifier needs based on the schematics:
Engineer's amp 50W monoblock
Hammond 290FX or equivalent
640VCT @ 200mA DC
60V mA DC bias tap
6.3V @ 5.5A filament
1) Forum members have advised me that because it's specified as DC current, 640VCT @ 300mA would be advisable
2) I received this through private message of a respected board member:
"When you spec a custom transformer usually the winder wants to know AC RMS current, not the DC B+ current. There are lots of rules of thumb. But in general you spec considerably higher RMS current than the DC current - as current is actually drawn in pulses. There are guidelines anywhere between 1.5x and 2x. I usually use 1.8x to 2x.
You are best off overspecing it. Specifying higher current will increase the cost a bit, but it will work better and run cooler.
I would probably spec 400mA RMS."
So far so good. I now understand the 300mA forum members were advising me to get. So i tried asking a custom winder for a quote:
640VCT @ 400mA
Bias tap at 60V
6.3V @ 5.5A
21V @ 1.7A (21v sweep tubes)
20V @ .2mA (typo, I should have specified 20V @ 0.12A for WE 408A)
3) This is where I get confused. One of the custom winders (will remain anonymous) replied with:
"The 640vct @ 400ma is a fairly large VA and will heat no matter, to have it
potted and you not knowing its actual loading is a disaster waiting to
happen. Having all those windings will assure its overheating.
Many multi-secondary transformers are a product of progressive design/s, cut
an try to work right.
To just design with a guess is what we do not do anymore.
Your request should be made into two or three separate transformers due to
heating and winding coupling distortion/s. best would be one HV trans and
one 6.3v/5.5A -21v/1.7A and one 60 and 20v bias whatever.
Also the RMS value is .707 of a peak value and has nothing to do with power
supplies except determining the wire size of a HV secondary. for 400ma use
.282 CM/ampere wire. We use this factor.
The actual running voltage you will get, is determined by your rectifier/s,
filter type, values used and idle and peak loading. Doubling the current
rating is not a real cure, it just makes the transformer bigger."
"Not much we can do for you, this is not buildable as stated."
What the amplifier needs based on the schematics:
Engineer's amp 50W monoblock
Hammond 290FX or equivalent
640VCT @ 200mA DC
60V mA DC bias tap
6.3V @ 5.5A filament
1) Forum members have advised me that because it's specified as DC current, 640VCT @ 300mA would be advisable
2) I received this through private message of a respected board member:
"When you spec a custom transformer usually the winder wants to know AC RMS current, not the DC B+ current. There are lots of rules of thumb. But in general you spec considerably higher RMS current than the DC current - as current is actually drawn in pulses. There are guidelines anywhere between 1.5x and 2x. I usually use 1.8x to 2x.
You are best off overspecing it. Specifying higher current will increase the cost a bit, but it will work better and run cooler.
I would probably spec 400mA RMS."
So far so good. I now understand the 300mA forum members were advising me to get. So i tried asking a custom winder for a quote:
640VCT @ 400mA
Bias tap at 60V
6.3V @ 5.5A
21V @ 1.7A (21v sweep tubes)
20V @ .2mA (typo, I should have specified 20V @ 0.12A for WE 408A)
3) This is where I get confused. One of the custom winders (will remain anonymous) replied with:
"The 640vct @ 400ma is a fairly large VA and will heat no matter, to have it
potted and you not knowing its actual loading is a disaster waiting to
happen. Having all those windings will assure its overheating.
Many multi-secondary transformers are a product of progressive design/s, cut
an try to work right.
To just design with a guess is what we do not do anymore.
Your request should be made into two or three separate transformers due to
heating and winding coupling distortion/s. best would be one HV trans and
one 6.3v/5.5A -21v/1.7A and one 60 and 20v bias whatever.
Also the RMS value is .707 of a peak value and has nothing to do with power
supplies except determining the wire size of a HV secondary. for 400ma use
.282 CM/ampere wire. We use this factor.
The actual running voltage you will get, is determined by your rectifier/s,
filter type, values used and idle and peak loading. Doubling the current
rating is not a real cure, it just makes the transformer bigger."
"Not much we can do for you, this is not buildable as stated."
Last edited:
Basically I'm wondering if there is truth to what the winder told me. And if anyone can look at the schematic and help me out. I guess lowering the mA rating might be somewhat beneficial...
If this winder speaks the truth would ordering this from another winder who agrees be a mistake? Overheating, inconsistent voltage? I honestly didn't think ordering a custom power transformer could be this complicated....And I think there other winders offer up to 12 secondary taps as standard.
640VCT @ 340mA
Bias tap at 60V
6.3V @ 5.5A
21V @ 1.7A (21v sweep tubes)Q
20V @ .15mA
If this winder speaks the truth would ordering this from another winder who agrees be a mistake? Overheating, inconsistent voltage? I honestly didn't think ordering a custom power transformer could be this complicated....And I think there other winders offer up to 12 secondary taps as standard.
640VCT @ 340mA
Bias tap at 60V
6.3V @ 5.5A
21V @ 1.7A (21v sweep tubes)Q
20V @ .15mA
Last edited:
I can highly recommend Heyboer, I do business with them frequently - all for powering tube audio electronics.
I am of the opinion that segregating some windings to a separate transformer can be helpful particularly if you want to sequence things.
I have had transformers made in the past that had at least 5 secondaries on them for filament power, and others with 3 - 4 moderate to very high voltage secondaries - all made by professional winders. None of those have failed in the past 20 years.
I am of the opinion that segregating some windings to a separate transformer can be helpful particularly if you want to sequence things.
I have had transformers made in the past that had at least 5 secondaries on them for filament power, and others with 3 - 4 moderate to very high voltage secondaries - all made by professional winders. None of those have failed in the past 20 years.
Well kinda. I have a pair of Tribute output transformers so it's a TOTL build for me. When I get my 3d printed CNC going I am planning something special for the case as well.
@kevinkr I'll check whether Heyboer will pot theirs. Also, your experiences on multiple windings eases my worry. Thanks
Last edited:
Yeah. Mine are probably the last that Pieter T wound for the DIY crowd. He is basically retired now hence the alternative winders.
Center-tapped transformers are typically rate for DC load current - it's pretty close to the RMS current in the winding anyway. Derating never hurts, but 2X is unnecessary. With a Hammond, get a 300 series for more iron and copper (they're 50 Hz rated) since it will run at a lower flux density. Lower core loss, less magnetic field.
Multiple windings DO complicate things - and it's never a good idea to have a winding with only a few VA rating as it can't be protected by the AC fuse - it'll burn up if there's a short.
If you are mass-producing something, you can go through a few prototypes to optimize a custom transformer. If you only want one, you need to specify EXACTLY what you need, as there's only one shot at it. Buy a standard part and it's already optimized.
Multiple windings DO complicate things - and it's never a good idea to have a winding with only a few VA rating as it can't be protected by the AC fuse - it'll burn up if there's a short.
If you are mass-producing something, you can go through a few prototypes to optimize a custom transformer. If you only want one, you need to specify EXACTLY what you need, as there's only one shot at it. Buy a standard part and it's already optimized.
What's the minimum VA rating for a winding to be protected by the fuse?
In other news I have narrowed down my search to Silk or Toroidy. Toroidy won't match cosmetically but can be placed beneath the chassis. Then a 3d printed cover can be made and spray painted to match.
At 400mA (I'll get 320VA)
AE - $280 each
Monolith - $270 each
Silk - $240/260 each (depending on lead style)
Toroidy - $170 each (Supreme)
What the quotes above do not give are shipping. Something that makes Toroidy more alluring.
Also, it occurred to me. In a country as large as the USA not one manufacturer pots theirs. That option is simple marketing. How hard is it to look at McIntosh & Citation II and notice they're sexy?
Tango (Japan) - potted
Tamura (Japan) - potted
Hashimoto (Japan) - potted
Softone (Japan) - potted
Noguchi (Japan) - potted (some)
Silk (Thailand) - potted
Toroidy (Poland) - potted
Lundahl (Sweden) - Optional case
Tribute (Holland) - potted
AE Europe (Holland) - potted
Amplimo (Holland) - potted
Monolith (Belgium) - potted
Plitron (America Jr.
) - pottedEverything USA - End bells
Last edited:
Here's an example:
Lets say we have a 300 VA transformer that also has a 10 VA secondary. Since the winding resistance might be 5% of the load (corresponding to 5% regulation), the short circuit current will be no higher higher than 20 times the normal load. So we have 100-200 watts being dissipated in the transformer winding, with as much as 200 VA additional input. If the transformer is normally loaded at perhaps 200 VA and fused at 3A (360 VA) it may never blow with a 400VA load - a 3A fuse can carry 4A (480 VA) for up to an hour. A 3A slow-blow will typically take one minute to open at 5A, 15 minutes typical, 1 hour max at 4A. Meanwhile the transformer has cooked.
Looking at it another way, if no secondary is less than 10% of the total rating, any winding short will draw at least twice the full load current, even if there are no other loads. Every situation is different, circuit protection takes some work.
I had Toroidy make me a PT for the DCPP.
I asked originally for:
TS200VA
PRI:240V
SEC:
280 - 50 - 0 - 50 - 280V (280mA)
3,15 - 0 - 3,15V (6A)
outer diameter: ~117mm / height ~47mm
But got him to increase the heater winding to 7.5A based on some posts on here. My voltages are very close to Pete's apart from the CCS varying things a bit.
The spec'd Edcor was getting super hot and buzzed a bit too. The Toroidy is quiet and doesn't get as hot - still warm though.
Last edited:
So Silk warned me that the shipping costs would be very high so that's out. I was going to go with Toroidy but every email was a different person replying. I doubt I would have gotten what I ordered so a new request if I go that route.
Then I reached out to Heyboer for a quote. Specified very low noise and well shielded. So this has come down to toroid vs EI power transformers.
Toroidy
+potted
+double coated copper electrostatic shield between primary and secondary (reduces mains line noise)
+magnetic shield (reduces stray fields)
+toroids have good regulation
+very low stray fields
+efficiency
-low DC tolerance
-higher mains line noise, high frequency hash
Custom EI transformer
+flux band
+lower mains line noise, high frequency hash is less
+Grain oriented core laminations (for higher permeability and less stray fields)
+Dual screen windings (The first one, which is closer to the primary is directly connected to the core, which should have an electrical connection to the chassis and thus to mains earth. The second screen is meant to be connected to signal ground. This provides an even improved isolation from mains noise.)
+External flux band - (To reduce stray fields)
+Vacuum varnished
-higher stray field
-less efficiency
Questions I have:
Which is more important for a tube amp. Lower mains line noise or lower stray fields?
How effective is an EI transformer flux band vs a toroid with or without a magnetic shield?
How effective is an electrostatic shield on a toroid vs a EI transformer with or without an electrostatic shield?
Then I reached out to Heyboer for a quote. Specified very low noise and well shielded. So this has come down to toroid vs EI power transformers.
Toroidy
+potted
+double coated copper electrostatic shield between primary and secondary (reduces mains line noise)
+magnetic shield (reduces stray fields)
+toroids have good regulation
+very low stray fields
+efficiency
-low DC tolerance
-higher mains line noise, high frequency hash
Custom EI transformer
+flux band
+lower mains line noise, high frequency hash is less
+Grain oriented core laminations (for higher permeability and less stray fields)
+Dual screen windings (The first one, which is closer to the primary is directly connected to the core, which should have an electrical connection to the chassis and thus to mains earth. The second screen is meant to be connected to signal ground. This provides an even improved isolation from mains noise.)
+External flux band - (To reduce stray fields)
+Vacuum varnished
-higher stray field
-less efficiency
Questions I have:
Which is more important for a tube amp. Lower mains line noise or lower stray fields?
How effective is an EI transformer flux band vs a toroid with or without a magnetic shield?
How effective is an electrostatic shield on a toroid vs a EI transformer with or without an electrostatic shield?
Last edited:
That's good to know but I wasn't doubting their quality. Just how a specific core type fits into my design.
Dual screen windings (The first one, which is closer to the primary is directly connected to the core, which should have an electrical connection to the chassis and thus to mains earth. The second screen is meant to be connected to signal ground. This provides an even improved isolation from mains noise.)
I specified the first electrostatic screen to have an electrical connection to the chassis and thus to mains earth. I did this because this is how Thomas Mayer specifies his and I am using Tribute Transformers so I do want quality. VinylSavor: Power Transformers
My question is how does this affect my build, before I submit payment? Do I need additional steps to ground the chassis? If so how? Do I connect the chassis to the second ground in the attached photo?
Although it sounds as though it's more than they can handle.. did they understand it was for full-wave rectification?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.