Evening all,
I've been embarking on more elaborate power supply designs, while going through Thomas Mayer's blog bit by bit. I'm wanting to include in my next build a choke input filament supply such as these examples, from his blog:
So I'm shopping up the components that aren't just on the shelf, and that means just the high current chokes... I won't be using 220, and it will be 5v from the transformer to 2.5v filaments @3amps for two 45 tubes.
The questions I have are:
1. What is the motivation to use a high-quality choke in this position, eg a $150 Lundahl? Why would I not use an $8 toroidal from mouser with the same inductance and current ratings? Is wire gauge important? Why wouldn't we all just go out back to that dead washing machine in our yards, pull the motor and unravel the wire, fetch a nice stick from the woods and wrap ourselves up the medieval clay-pot battery-accessory of an inductor that we all can have for free?
2. Will precise inductance matter much at all in this position? As long as DCR is right, am I correct that I can use a wide range - and would it be correct that I can go ahead and stick to the low end of that range, to minimize size without impacting the function? Say, 1 to 10 mHy?
Thank you in advance for any information. Links are quality input in my book.
I've been embarking on more elaborate power supply designs, while going through Thomas Mayer's blog bit by bit. I'm wanting to include in my next build a choke input filament supply such as these examples, from his blog:
So I'm shopping up the components that aren't just on the shelf, and that means just the high current chokes... I won't be using 220, and it will be 5v from the transformer to 2.5v filaments @3amps for two 45 tubes.
The questions I have are:
1. What is the motivation to use a high-quality choke in this position, eg a $150 Lundahl? Why would I not use an $8 toroidal from mouser with the same inductance and current ratings? Is wire gauge important? Why wouldn't we all just go out back to that dead washing machine in our yards, pull the motor and unravel the wire, fetch a nice stick from the woods and wrap ourselves up the medieval clay-pot battery-accessory of an inductor that we all can have for free?
2. Will precise inductance matter much at all in this position? As long as DCR is right, am I correct that I can use a wide range - and would it be correct that I can go ahead and stick to the low end of that range, to minimize size without impacting the function? Say, 1 to 10 mHy?
Thank you in advance for any information. Links are quality input in my book.
Maybe this is usefull for question 2: ARRL Radio Amateur Handbook 1968 : Free Download, Borrow, and Streaming : Internet Archive
As far as I can see, the supply works with a swinging choke. The inductance of a swinging choke has to be higher than a certain critical value for it to function properly (below the critical value the output voltage would be quit a bit higher). The link provides the formula to calculate that critical value, although my knowledge doesn't go deep enough to be sure if the formula also applies 'one-on-one' on these low voltages/high currents.
If I'm not mistaken, the construction of a swinging choke differs a bit from that of a normal choke. The swinging choke has a smaller air-gap since it has to 'saturate to swing'.
So with a swinging choke, the right output voltage is not just a matter of getting the DCR right.
With the formula, I calculated a minimum inductance of 3.1 mH for the 211. But higher values will reduce the ripple more. I guess that's why the choke in the schematic with the 211 is 0,1 H.
Addition/correction: Because I don't know the resistance of the second choke, I can't know the voltage after the first choke. I calculated with 10 V but that should be higher because of the voltage drop over the second choke.
As far as I can see, the supply works with a swinging choke. The inductance of a swinging choke has to be higher than a certain critical value for it to function properly (below the critical value the output voltage would be quit a bit higher). The link provides the formula to calculate that critical value, although my knowledge doesn't go deep enough to be sure if the formula also applies 'one-on-one' on these low voltages/high currents.
If I'm not mistaken, the construction of a swinging choke differs a bit from that of a normal choke. The swinging choke has a smaller air-gap since it has to 'saturate to swing'.
So with a swinging choke, the right output voltage is not just a matter of getting the DCR right.
With the formula, I calculated a minimum inductance of 3.1 mH for the 211. But higher values will reduce the ripple more. I guess that's why the choke in the schematic with the 211 is 0,1 H.
Addition/correction: Because I don't know the resistance of the second choke, I can't know the voltage after the first choke. I calculated with 10 V but that should be higher because of the voltage drop over the second choke.
Last edited:
I’m sorry… but I don’t see the issue. Rectify a 12 volt secondary. Lose something around 2 * 0.8 V per diode, so 1.5 volts. Leaving 10.5 RMS Into a large enough choke, the output will be 0.8 * RMS or 8.4 VDC. This is not optimal for your 6.3 VDC service.
The really simple solution there is to choose ‘lossier’ rectifier diodes. Higher VF at a nominal 1 amp, say. BY822 Vishal, buck a throw, 1.5 VF at 1 amp. The quad of them removes another 1.5, delivering 6 to 7 VDC.
Rather than that though, why not just do the same rectification, into a modest sized (1000 uF) cap, a modest (as shown) L, then an LDO 6 V regulator? WAY better ripple control, cheaper than dirt, and solid load independent voltage level control. Dunno… I feel the marriage of top-shelf semi and vacuum is righteous.
My opinion of course.
GoatGuy
The really simple solution there is to choose ‘lossier’ rectifier diodes. Higher VF at a nominal 1 amp, say. BY822 Vishal, buck a throw, 1.5 VF at 1 amp. The quad of them removes another 1.5, delivering 6 to 7 VDC.
Rather than that though, why not just do the same rectification, into a modest sized (1000 uF) cap, a modest (as shown) L, then an LDO 6 V regulator? WAY better ripple control, cheaper than dirt, and solid load independent voltage level control. Dunno… I feel the marriage of top-shelf semi and vacuum is righteous.
My opinion of course.
GoatGuy
Last edited:
Thank you both, I think I'm going to grab an affordable choke from fleabay, labeled 40 mhy 6 amps (wow but they are affordable still), and give LC a shot. I'll get a few different diodes to try, and get that voltage as close as possible before trimming it on the other end.
I think I'll also pick up a transistor supply and compare the results. This would be a useless effort if I'm unaware of the alternative.
I think I'll also pick up a transistor supply and compare the results. This would be a useless effort if I'm unaware of the alternative.
I've always thought that those old fashioned ballast chokes for lights would be excellent input chokes for filament supplies; the highest current ones are about 0.5 H.
That's the kind of thing that gets me going, I'm a junkyard dog. What I've found is that chokes that can handle the amperage tend to be 50 or more anyways, because they're really huge. I was looking at one on the bays that was $20 but was quoted at $60 shipping lol. We are talking 10lb units. It would be pretty hilarious to dwarf the hv transformer with the filament chokes no? What a hobby.
Here's a few chunkers that I'm looking at:
SCHAFFNER RD8137-16-5m0 CURRENT-COMPENSATED CHOKE,600VAC/850VDC, DC to 400 Hz | eBay
TODD ELECTRIC FILTER CHOKE 40.0 MH 6 Amps .35 Ohms 50-400 Cycles Cat No. CH-06B | eBay
Here's a few chunkers that I'm looking at:
SCHAFFNER RD8137-16-5m0 CURRENT-COMPENSATED CHOKE,600VAC/850VDC, DC to 400 Hz | eBay
TODD ELECTRIC FILTER CHOKE 40.0 MH 6 Amps .35 Ohms 50-400 Cycles Cat No. CH-06B | eBay
Last edited by a moderator: