Choke Voltage Rating

[defaced]

Is a choke's max voltage rating with respect to ground, or is it with respect to the voltage drop across the choke?

[tbryanh]

Assuming the core is grounded, it would probably be with respect to ground.

The 2 ends of the coil can be shorted together and connected to a voltage above the maximum, and arcing to ground would probably occur. This would happen even though zero volts is across the coil.

If the choke had an air core, the maximum instantaneous voltage across the coil could probably be much higher than with an iron core. (The voltage would divide among the turns, so there would be less chance of arcing across the turns.)

[defaced]

Hmm, so if I isolate the core from ground, say by using a painted chassis and nylon screws, can I get away with using a 400v inductor with say a 550v B+?

[tbryanh]

It appears an isolated core allows a coil to withstand higher voltages, but does an isolated core inject noise electrostatically or electromagnetically to nearby components? There is probably some reason the core is grounded.

[tedmich]

don't these saturate at too high a voltage?

[mooreamps]

There is probably some reason the core is grounded.

Be-cause . it's . made . of . metal ? ? ?

Hmm, so if I isolate the core from ground, say by using a painted chassis and nylon screws, can I get away with using a 400v inductor with say a 550v B+?

Yes. I do. Keep in mind, those nylon screws can not handle more than a pound of weight, or so. They will not support a heavy high henry choke. However, you don't need a high henry choke ; if you tank it at 120 cycles.



-g

[dai h.]

mike, here's some info from an old thread(here) that was on my hard drive which has some info on the voltage rating:

START NEW THREAD REPLY Top of Thread LIST
Read 152 times
From: JWK @
Date: 3/15/2002 7:45 PM
Subject: Voltage rating for input chokes


I need to know how the voltage rating is determined for an input choke. I've decided to use a different transformer for my SE KT88 amp because the 400-0-400 tranny I have with an input choke is not going to give me the voltage I want. According to the Hammond site I would be getting around 325VDC with an input choke. I would like closer to 400VDC.

So with the center tap transformer like these, do you multiply by 1.41 to get what the choke "sees"? Ie., a 500-0-500 tranny would give the choke 500(1.41)=705. That would mean one would need a choke rated for 800 volts. Is this right?

Also, I thought to get the VDC from an input choke, you multiplied by 0.9

So a 400-0-400 would give you 400(0.9)=360

Hammon gives a 425-0-425 tranny giving 350VDC. 425(0.9)=382. Maybe that's without a load, but I would think 32 volts is an awful lot to drop. Maybe that's what an amp with four EL34's would drop. Maybe a SE amp would only drop 10 volts, especially with a monster PT like the one I have. Just guessing, but I don't know so please someone tell me what's real. Thanks.

John

START NEW THREAD REPLY PREVIOUS LIST
Read 136 times
From: JWK @
Date: 3/17/2002 12:31 AM
Subject: Re: Voltage rating for input chokes



bttt

START NEW THREAD REPLY PREVIOUS LIST
Read 128 times
From: Sean K
Date: 3/17/2002 9:24 AM
Subject: Re: Voltage rating for input chokes



Man,I know how you feel.Have you got the PSUD from duncan amps?
http://www.duncanamps.com/software.html
This way you can tap in all sorts and see what happens.

START NEW THREAD REPLY PREVIOUS LIST
Read 118 times
From: JWK @
Date: 3/17/2002 4:59 PM
Subject: Re: Voltage rating for input chokes



Yes, I have that program and confess willingly that I am too ignorant to use it correctly. I just kinda stumble around with it and get frustrated. There's just too much I don't know.

In any case, the program assumes you know how to rate your choke correctly regarding voltage.

I was mostly asking for the sake of my own knowledge and for my future SE KT88 amp. I have the OPT for that, some JJ KT88 tubes and various components, but I'm in the middle of building my very first amp, so I'm going to let this one go for now and come back and harrass everyone after my first one is done.

John

START NEW THREAD REPLY PREVIOUS LIST
Read 93 times
From: Sean K
Date: 3/18/2002 10:12 AM
Subject: Re: Voltage rating for input chokes



I can sympathise.I'm stumbling along myself with this.What are the values for the choke and waht size cap are you going to follow it with and what sort of rectification are you going to use and I'll go and play with the designer.

START NEW THREAD REPLY PREVIOUS LIST
Read 83 times
From: Wild Bill @
Date: 3/18/2002 5:59 PM
Subject: Re: Voltage rating for input chokes



John, that 1.41 is the factor of peak to RMS voltage. You use it with a filter cap because although you apply RMS the cap will charge to the peak value and once it's "full" it will stay there. Of course, as soon as you begin drawing power into a load the voltage drops big time, because the transformer is not re-supplying steady even power but rather AC pulses. So the voltage in the cap will tend to "average down" as more power is drawn from it.

Of course, you hit a point where the voltage won't drop anymore. This is the average AC value, which is .9 RMS.

A choke input filter is different because there IS NO cap first in the line! This means the voltage is building energy in a magnetic field, like a capacitor charging to peaks and releasing during the dips in the cycle. But now we're talking CURRENT, not voltage like with a cap! We put a filter cap AFTER the choke and let the "choke-ed" voltage supply it.

The voltage on this cap will be .9 RMS, steady as a rock. So a 500-0-500 winding in CT half-wave will supply 450 VDC.

Just like with a cap, chokes need a minimum value to be useful. Even more, they need a minimum value to initially build up their magnetic field and hold it. Since the field comes from current flow it needs a minimum "holding current" to work. Below this point the field either totally collapses or is just too weak to do its job. We calculate minimum inductance like so:

L=V/I(ma) , or inductance equals the voltage divided by the current in milliamps.

So if you had 500 volts and an idle current of 100 ma you need a choke of 5 Henries (or bigger!) to work like a choke input filter.

What happens if the current gets too low? The choke no longer acts like a choke and is just a "resistor" of the copper resistance of the wire in its winding, usually a few hundred ohms or less. So when it no longer does anything useful you now essentially have a CAP input filter and the voltage on the cap rises to - 1.41 RMS! This can get a little too exciting if the cap can't handle the new voltage...

So it's important to maintain that critical inductance. It's easy with cathode biased amps - as they warm up they start to draw idle current. A tube rectifier that warms up with everything else is nice as well. With SS diodes as a rectifier they instantly deliver B+ before the tubes start to draw current so you'll get that high voltage on the cap after the choke unless you allow for it in your circuit. Bleeder resistor(s) can help and are always a good idea anyway to drain off the B+ when the amp is shut off and you're gonna poke your fingers inside it! Or just make sure the caps can handle the voltage anyway.

I'm running the risk of getting flamed for not being totally scientific in my explanation here but as a working "mind-picture" of what's going on I think it 's practical.

Also, when you're worrying about the voltage rating of a choke remember there's TWO ratings! The first is for the voltage across the choke. This is usually not a problem because in your amp power supply this is the drop from plate to screen voltage. Not really a lot. The other is the voltage rating to ground, so the winding doesn't zap out to the iron. Not a problem with most chokes in most amps but if you're running a monster you can put the choke in the negative CT line instead of "on top" and it'll see a low voltage to ground but work just the same!

I've rambled on but I hope this helps you...

---Wild Bill

START NEW THREAD REPLY PREVIOUS LIST
Read 65 times
From: Sean K
Date: 3/18/2002 11:14 PM
Subject: Re: Voltage rating for input chokes



Thanks Wild Bill,
Newbies like me and John,I would suppose(I don't want to be wrong on anything either)
often have trouble with the scientific answers that suppose we already know everything and feel very relieved when pictures are offered as explanations.
I've asked some questions,in the past,and been thoroughly blitzed with science and gone DUH,then found a simple drawing somewhere or an explanation
and suddenly gone yeah,and as I then get it the science starts to be useful.
A classroom works well when the slightly more advanced students help the less so and this gives more room for the "teacher" to be more effective but if this isn't happening and the teacher goes around "flaming" anybody helping out then were going backwards.
I wrote out several explanations to the previous,not up to yours,but a little beyond Johns,but scrapped them 'cause I was afraid of teacher!,In high voltage enviroments this is debatably a good or a bad thing but judging by the amount of reading of posts going on theirs a whole bunch of people holding back for maybe this reason.
Not a reason for change but just some observations
and I hope I haven't thoroughly bored anybody,Sorry,I won't do it again,promise,please don't hit me!

START NEW THREAD REPLY PREVIOUS LIST
Read 62 times
From: JWK @
Date: 3/19/2002 2:08 AM
Subject: Re: Voltage rating for input chokes



Thanks for the reply and info. Some of that I got from KG a while back. I printed out that reply and he had given me the formula for the minimum current and explanation that agrees with yours 100%.

The new tentative plan is to use the Hammond 279X PT which is 425-0-425 rated at 150mA. This will give me a bit higher voltage than the 400-0-400 I have and won't be anywhere near as rediculously large. The choke I would like to use is the 93D which is 8 henries rated for 150mA with a maximum working voltage of 600 volts. And there's the rub. I don't know what Hammond means by "maximum working voltage".

So I have two issues to resolve. The first is that I would like to experiment with inexpensive radial caps after the choke that can be rated for 400v or 450v. It becomes a pain and twice as expensive to series two together to insure against the 600 volts that the cap would see if there is not enough current through the input choke. Soooo, how would one do this? Put the standby switch just after the choke and before the first cap? With all the filaments heated, wouldn't there be current immediately upon throwing that switch? The second would be to know if that choke can handle the "working voltage" of this particular setup.

An aside question would be why the voltages given on the Hammond web site are so low for input chokes?

John

START NEW THREAD REPLY PREVIOUS LIST
Read 64 times
From: Wild Bill @
Date: 3/19/2002 3:11 AM
Subject: Re: Voltage rating for input chokes



John, 600 volts is probably to ground. The way Hammond over rates their products, I'd say just go with it.

How much idle current will the KT 88 be drawing? This will let you calculate your critical minimum inductance. If you put a bleeder resistor of 200k or so at 5 w to be safe you'll pull a few more ma. on startup.

Putting the standby switch before the first cap sounds like a good idea but what about when (not if!) someone doesn't know or forgets? I only see one easy out against having 600 volts of cap - a tube rectifier with a warmup time. What's wrong with a tube rectifier anyway?

Maybe someone else has an idea to let you keep using SS rectifiers...

---Wild Bill

START NEW THREAD REPLY PREVIOUS LIST
Read 55 times
From: JWK @
Date: 3/19/2002 5:07 AM
Subject: Re: Voltage rating for input chokes



The KT88 will be drawing somewhere around 100mA, so the minimum inductance would be 3.8H with the 425-0-425 tranny. Is that right?

I here you about the "when" and not "if". I was thinking more in terms of finding the uF value I liked best for the first cap without the hassle. I suppose once I found that value, doubling up some 350v caps would be the thing to do.

As far as the tube rectifier, there's nothing wrong with them. I like them. They make things simple. They sound good. BUT I don't want to drop any voltage AND it was my understanding (perhaps wrong) that a 425-0-425 transformer is beyond the voltage limit of a 5AR4/GZ34. Aside from the voltage being too high for the rectifier tube, I'm thinking that 385 volts would drop down to 365 volts with the forward voltage drop, then lose another 15 to 20 volts from the load. Or maybe that rock solid voltage from the input choke wouldn't budge with the amp's load? Say now, there's a thought...

Soooo, can I use a rectifier tube and how much voltage will I lose? Thanks for the help.

John

START NEW THREAD REPLY PREVIOUS LIST
Read 61 times
From: kg @
Date: 3/19/2002 1:41 PM
Subject: Re: Voltage rating for input chokes



if you do go with a tube rec, keep in mind that when you use a choke input filter the filter caps no longer have a restriction on the maximum size permissible. this is because the conduction current is limited by the inductance of the choke--you could hang 1,000 uF of caps off the b+ rail and the tube rec would be perfectly content.

tube recs also require no snubbers to keep electrically quiet.

ken

START NEW THREAD REPLY PREVIOUS LIST
Read 46 times
From: JWK @
Date: 3/19/2002 4:49 PM
Subject: Deciphering the RCA manual



I'm looking at the rating chart in the RCA manual. If I'm reading this correctly, it looks like the maximum voltage per plate would be 600v @ 100mA for a 5AR4. Rectifier tubes have always mystified me, even more so than the other stuff I hardly know about. Looking over the charts, it would seem the voltage limit is somewhat dictated by the amount of current running through the tube. Just like any other tube, I guess.

In any case, what I can see is that with an input choke for a SE amp set at 100mA idle, A 5AR4 or a 5U4 would work just fine with a 500-0-500 PT, where a 425-0-425 PT for a typical push pull EL34 amp would fry it in no time. Is this correct?

Thanks to all of you for helping me with this basic stuff. I still don't want to make a clone.

John

START NEW THREAD REPLY PREVIOUS LIST
Read 25 times
From: Wild Bill @
Date: 3/20/2002 4:39 AM
Subject: Re: Deciphering the RCA manual



John, you use 2 plates, one for each half of the cycle. So your current rating is 200 ma.

Tube rectifiers are like SS diodes - they have a PIV rating and a current rating. They also have a voltage drop that increases with current drawn. Unfortunately, many data sheets don't mention it.

---Wild Bill

[tbryanh]

don't these saturate at too high a voltage?

Too much current causes the core to saturate.

[tbryanh]

There is probably some reason the core is grounded.

Be-cause . it's . made . of . metal ? ? ?

So its just a no brainer? Ground anything that is metal?

[black_labb]

good discussion above dai

i assume the formula in the following is for VAC (rms) from the tranny into the choke? useful information.

"Just like with a cap, chokes need a minimum value to be useful. Even more, they need a minimum value to initially build up their magnetic field and hold it. Since the field comes from current flow it needs a minimum "holding current" to work. Below this point the field either totally collapses or is just too weak to do its job. We calculate minimum inductance like so:

L=V/I(ma) , or inductance equals the voltage divided by the current in milliamps.

So if you had 500 volts and an idle current of 100 ma you need a choke of 5 Henries (or bigger!) to work like a choke input filter."

[dai h.]

hi, sorry I'm am pretty pathetic at math so I'm not much use when it comes to numbers but 500/0.1=5000 and 5000 looks much bigger than 5!

the only thing I can add as far as chokes is that I think (if someone is going for a choke input supply) the choke needs to be designed for the purpose and if it isn't up to the job it will vibrate and hum (as I understand). The ones I've seen advertised as choke input supply capable also seemed more expensive.

[black_labb]

notice that the formula notes milliamps, as opposed to amps.

[dai h.]

ahh, thanks so 500/100=5. I see how the 5 was arrived at now.

[Steve Conner]

I've used tube amp chokes with about 7kV RMS at 400Hz across them. I used four identical units in series, isolated from the chassis, with each having its core connected to one end of its winding. So each choke saw about 1.7kV RMS across it.

All four chokes were still good when I dismantled this project, and I reused them in tube amps where they are still working now.

I concluded that the voltage rating is probably just an ass-covering legal move by the manufacturer.

What will cause the choke to saturate is current, and the current can be worked out from the voltage across the winding as I= (1/L)*(time integral of voltage). In a capacitor input filter, the current ripple in the choke is small, so it will probably overheat before it saturates.

In a choke input filter, the current ripple is large, so you have to worry about saturation on peaks. Nobody uses choke-input filters any more (they were a band-aid to get round the low peak current rating of tube rectifiers) and I'm not sure anyone sells chokes designed for them.

Also, metal things only need to be grounded if they're somewhere that people can touch them. For the sake of a service engineer, though, live metalwork inside equipment should be clearly labelled.

[defaced]

Isolate the choke it is. Thanks for the notes on practical experience.

Would that formula (L=V/I) be applicable for a choke in a Pi filter? Running my numbers, 550v and 50ma, I'll need an 11H choke. So similar to the first question, is the voltage in this equation with respect to ground or with respect to the voltage across the choke?