Finding spec of unknown choke.

[Diabolical Artificer]

I have several choke's of unknown inductance, voltage and current rating and would like to find these before using them.

Having looked back at threads on the subject of finding a coil's inductance, I've seen that by putting a cap in parallel we can find it's resonance and thus work it's inductance. I've tried this
by putting a .1u cap in parallel with choke X. Started off at 20khz at 2v but at 50hz the amplitude still wasn't half.

Reading further I see we need to put some DC in to get a valid result and can use the attached circuit. So if I have this right if I want a choke to filter 450v HT i set up the test circuit at this voltage, but what about the chokes current rating?

Choke X in this case is roughly 4 x 4 x 3 inch's with a stack 8cm x 4cm (more accurate with my ruler to measure in cm) the copper wire is 73 thou thick.

One other concern is how do I know it will handle X volt's without destructive testing?

TFL, Andy.

[John M0GLN]

Is the wire diameter 0.073"?, If so it's slightly bigger than 16SWG and my F J Camm Wireless book lists it OK for 6.5A max for a transformer.

John

[Ed_Dinning]

Hi Andy, it can get a bit complex with smoothing chokes.
They normally have a low signal level (no DC) inductance value. If the signal is bigger and especially if DC is present this can reduce by a lot.
There are some nice (and expensive ) bridges that will let you measure this effect.
The current rating is normally based on the wire rating of 3A per mm^2. If you cannot easily measure the wire dia, run the choke on low voltage DC at the current you need and see how hot it gets. If more than 50C on the outside, it is a lower current than you thought.
Note that DC current is not only the heating effect, but will cause saturation and reduction of L value.

Ed

[ToneArm]

I bought a cheap component tester off ebay, just a chinese pcb without a case. Think it was about £15 but did take a couple of weeks to arrive. It claims it can measure inductance - not tried it but I'll give it a go this evening and report back.

Glyn

[ToneArm]

That was easy, it does indeed measure inductance. Good value gadget!

[PJL]

Wound with 16SWG, its not going to be suitable for use with a 50Hz HT supply where 10H would be typical. Unless that is a lead-out wire you are measuring. Try measuring the resistance.

[Diabolical Artificer]

The wire diameter I'm measuring is a lead out wire but it looks like the winding itself as it's enamelled copper. Been a while since I used a mic but it still fits on the little finger, I get two big division's @ 25thou each = 50 and 23 on the barrel so 0.073 is right.

I'll try connecting it up today Ed and see what I get, the DC res is 0.2 ohms. Like you, I think I have a Marconi TF2700 that measure's inductance but I've found to hard to measure unknown L's. As you know there is a add on box TM 6113 to measure power choke's but they're hard to find. There are two tuned circuit inside - see att, I suppose it would be possible to work out the value's.

On the whole then is a choke suitable for 50hz HT supply going to have thinner gauge wire and a higher DC res? I would make finding a suitable choke quicker if that was the case.

Thanks all, Andy.

[russell_w_b]

Put your choke in series with a low-value resistor and connect to a low-voltage a.c. supply. Measure and note the supply volts, the volts across the resistor, and the volts across the choke. You could use a 60W lamp as the resistor and the mains as a supply.

Use the Cosine rule to extrapolate the reactive and resistive (at a.c.) components and acquire an a.c. resistance and an inductance.

b = cos^ -1 [(A2 + C2 – B2 ) / 2AC]

Where A represents the voltage across the choke; B represents the supply voltage, C represents the voltage across the series resistor, and b represents the angle between the in-phase voltage across the series resistor and the voltage across the choke.

The resistive ‘in-phase’ voltage component is A Cos (180-b) and the inductive quadrature voltage component is A Sin (180-b). The current to acquire the a.c. resistance and the inductive reactance will be the voltage across the resistor divided by its resistance (or a series meter, if you have one spare).

This method will draw 'real' current through the choke and proffer an inductance (and a resistance at a.c.) commensurate with losses. The values obtained will differ considerably from those values measured on a signal-level impedance bridge.

Of course, a d.c. component will change things again...

[Skywave]

Originally Posted by Dr Wobble

On the whole then is a choke suitable for 50hz HT supply going to have thinner gauge wire and a higher DC resistance? I would make finding a suitable choke quicker if that was the case.

Presumably, when you say "50 Hz HT", you mean full-wave rectified a.c. whose freq. is 100 Hz.
However, just few comments, of which you may already be aware of, but I'll mention them just for the record.

Choke types: bear in mind that when it comes to 'smoothing chokes' for PSU applications, there are two types: nominal 'fixed' inductance and 'swinging chokes'. The latter are usually preferable in this application and usually result in a smaller sized item than the former.

Choke resistance: ideally, any choke or coil will have zero d.c. resistance: it's the inductance that is the required electrical quantity. Problem there, of course, is that for a required value of inductance, as the R goes down, the wire thickness increases, so the weight and size of the choke increases - and usually disproportionately.

Whenever I have built custom HV PSUs, practical limitations have usually dictated the choice.
1. How much space is available on the chassis for the choke?
2. What chokes are available in my store? (Usually salvaged items).
3. Choose a few with the lowest resistance but are also large and commensurate with limitation 1 above.
4. Build a quick test cct. to evaluate the overall design (including proposed mains transformer, filter & smoothing capacitors); hang various power Rs on the O/P, the values of which to simulate the intended load.
5. 'Soak test' the whole affair for, say, one hour. Apart for the power Rs, how hot does the transformer and choke get?
6. Draw the obvious conclusions; re-design if necessary.

Al.

[russell_w_b]

Originally Posted by russell_w_b

Use the Cosine rule to extrapolate the reactive and resistive (at a.c.) components and acquire an a.c. resistance and an inductance.

b = cos^ -1 [(A2 + C2 – B2 ) / 2AC]

That should read b = cos^ -1 [(A^2 + C^2 – B^2 ) / 2AC]

1. Voltage rating, it's only a coil on a former, even a badly made one would be good for a kilovolt or two.

2. Current rating, current squared x resistance = watts, for an iron cored thing a watt or two per cubic inch is fine.

I wouldn't worry about the inductance too much, most chokes of that era where designed down to a price for HT supplies, anything big enough will do.

[Herald1360]

A choke with 16SWG windings is unlikely to be for HT supply use where 100s of milliamps is more likely than a few amps, at any rate in radio/amplifier service.


In terms of things to watch out for with choke filters- the primary resonance of the choke with its capacitors needs to be watched out for. Get it too close to the ripple frequency or too high a Q at any frequency and you can get real trouble with voltage overshoots or an unstable PSU on a varying load.


HT PSU chokes are normally OK on account of their series resistance offering adequate damping but for higher current supplies it can be problematic.

[Ed_Dinning]

Hi Andy, resistance of coil is probably too low to be an HT choke.

Your TF2700 can be adapted to measure a choke with DC passing by duplicating the circuit in the adaptor box that is shown in the TF2700 booklet.
Using the TF2700 for chokes it is essential to adjust the power factor control at the same time as the main balance, a 2 handed job!

Ed

[Diabolical Artificer]

Connected said choke up as suggested Ed, slowly cranked the HT up 400v without any overheating. Scoped the both ends and the result was odd. I connected it across a 16/16u can, for testing and expected to see the usual sawtooth wave, but instead got a non repetitive wave which "jumped" every so often. DC remained solid but the DMM on AC the ripple was all over the shop.

As said by Ed and other's this is probably not a good candidate for smoothing HT supplies. I found another which has thinner gauge winding and a DC res of 2.5. It measured 330mH on the 2700.
Interestingly I could find the L of the first choke with the 2700. Using the TF2700 is a bit like cracking a safe, of two safe's simutaneously.

Bless you Russell, I was asleep or legging it out the school gates when we did cosine's. Our math's teacher had a drawling monotonic voice which could put you asleep in 3 second's. TBH I didn't want to learn then anyway, so I'm afraid a math's heavy approach is going to be tricky. I don't even know how to use a calculator for such sums.

Thanks Al, 100hz of coarse your right, slip of the ol brain there. I'll bear what you said in mind. There's a lot more to using a choke for a PSU than a RC filter. I wonder if there's much to be gain using one what with the extra weight. still it's about time I had a craic at using them.

Thanks Merlin and Chris. I'll do some more experimentation and report back.

Andy.

[trobbins]

Andy, as indicated, any sort of B+ power supply worthy choke for valve equipment is going to have pretty thin winding wire, and a DC resistance of probably at least 30-50 ohm, and easily up to 500 ohm or more. So that would save you some quizzical moments if you start eyeing off other likely candidates.

You should also aim to test down at low voltage levels - initial testing up at typical B+ seems like an accident waiting to happen - a modified test process similar to above suggestions is in link. Apart from DC resistance, a 'megger' / insulation resistance tester at 500 to 1kV from winding to core is also a good initial gating test - no point spending time on something that doesn't pass those two simple tests.

Tim
http://dalmura.com.au/projects/Choke%20measurement.pdf

[russell_w_b]

Originally Posted by Dr Wobble

Bless you Russell, I was asleep or legging it out the school gates when we did cosine's. Our math's teacher had a drawling monotonic voice which could put you asleep in 3 second's. TBH I didn't want to learn then anyway, so I'm afraid a math's heavy approach is going to be tricky. I don't even know how to use a calculator for such sums.

Oh, well... It's a useful dodge for extrapolating the resistive and reactive components from any sort of coil (chokes, telephone bells, solenoids, etc...) in the absence of a bridge. And, as it's done at 'real' current with nothing fancier than a decent multimeter or two, a lamp or resistor, and a mains or mains-derived a.c. supply, it factors in hysteresis and copper loss.

Granted, it's maybe not so useful in the case of a d.c. components, as here.

It's not too onerous to do, and most mobile phones - doesn't even have to be a smart-phone - with a 'calc' function can handle the sums. No need to buy a calculator!

[Diabolical Artificer]

Thanks Tim. I slowly upped the voltage with a variac, started at 50v then went up in 50v increment's.

I won't write off your advice Russell, it'll take me a while to get my head round it.

Andy.

[russell_w_b]

Originally Posted by Dr Wobble

I won't write off your advice Russell, it'll take me a while to get my head round it.

Good. Just whenever you've time to kill...

Actually, the information in the link provided by Tim (trobbins) is more useful in the context of the original post, as it factors in the d.c. component too. A good, useful read.

[Diabolical Artificer]

It seems all the choke's I have are unsuitable for smoothing filtering HT, apart from two I got off Al Astral Highway. I found some useful info here - http://livinginthepast-audioweb.co.u...p?p=vxformers2 which helped me to understand them a bit better.

Looking at commercialy available chokes it seems that the higher the power handling is the lower the inductance is, EG for a 1A ish choke 300mH.

I now have a lot better idea how to go about finding the specs of choke X, thanks for all your reply's.

Andy.

[Silicon]

The low resistance chokes may have come from a mains power supply adapter for a battery powered valve radio.

The directly heated valves were sensitive to mains ripple and so inductors were used in the smoothing circuit.

They rarely experienced a voltage above 12V so be careful when you 'Megger' test it.