Confirmation of DC constant current calibration in a bridge. - UK Vintage Radio Repair and Restoration Discussion Forum

mpegjohn · 8th Dec 2016, 11:09 am

Hi.

To test a choke I am winding I have built an Owens bridge, with a MOSFET constant current source.

To check the accuracy of the bridge, I have compared some chokes, against an Agilent LCR meter we have at work. But this does not have any DC current source.

The question in my mind is, is my current source messing with my measurement results.

So, I would love to know if anyone has a bridge or LCR meter with a CC source, that they know is good, that will allow me to send my choke, and do a few measurements so I can compare and contrast.

It doesn't need to be my choke either, I just need a reference piece, with known inductance at specified DC currents.

Any help will be appreciated.

The Owens bridge is a series of 4 articles here:-

http://diyaudioprojects.com/mirror/m...com/sbench101/

Regards, John.

trobbins · 8th Dec 2016, 11:18 am

John,

Do you need an accurate measurement of inductance at a given DC current, and what are your nominal levels (eg. mA, or A, mH or H, or ...) ? Or does your application allow some latitude - eg. if you were building a power supply filter choke?
Ciao, Tim

mpegjohn · 8th Dec 2016, 11:27 am

Tim,

I need an accurate reading of inductance at various DC levels. Inductance ~10H DC 0-100mA, for the inductors I have, but if not using mine, then any inductor > say 1H with Idc upto 200mA.

The actual inductor I am not interested in, just some form of cross checking my bridge setup.

John

PJL · 8th Dec 2016, 2:53 pm

How about measuring the 'inductance' of a resistor with the same resistance as the inductance you want to test over the current range?

mpegjohn · 8th Dec 2016, 3:27 pm

Not a bad idea at all.

mpegjohn · 8th Dec 2016, 7:50 pm

Just had a thought though, if I have a 10H choke at 100Hz that's 6K. at 100mA that will need at least 600V, my CC source has 110V max.
Maybe I could use Tim Robins method of choke measurement to get fixed points of DC and AC.

Argus25 · 10th Dec 2016, 2:43 pm

As you note, if you could get an accurate assessment of a sample inductance, by another method, then you could check your bridge calibration. Your Owen's bridge will probably be very accurate for small range inductances. Problems can crop up measuring large inductances though on many bridges and meters.

There is one way around the problem of large inductors and measuring or estimating their value. I have used this method in the past, it works:

If an experimental inductor is wound with much thicker wire and far fewer turns on the same bobbin and core, provided the physical space occupied by the winding on the bobbin is similar to the final high inductance version (usually the most efficient design uses all the available free cross sectional area of the bobbin) you can measure the much lower inductance value very easily and accurately on most inductance meters and bridges. And since the inductance is proportional to the square of the number of turns it is easy then to calculate the required number of turns for the final large inductor value you want. And the inductance should be correct, even if it is an extremely high value. Then you can check it with the bridge to see how accurate the bridge is in the higher ranges.

You can also check the effect of DC current on the lower inductance version, being mindful of the ampere-turns effect, which will be the same on the final large inductance so you can transpose the values because you will also know the turns numbers and the ratio of the DC resistances of the two inductor versions and any applied DC voltages.You will need a much lower DC voltage for the smaller value version of the inductor for this test.

So you could use that method to check the performance of your bridge for small vs high value inductors. For the low range values I would simply recommend comparing it to a commercial meter.

One problem with large inductors that have many turns of fine wire is that the self winding capacitance and coil DC resistance can become significant. The capacitance can foul up the reading on some meters & bridges depending on the operating frequency and principle of the measuring system.

One method of measurement is to tune them with a 0.1uF capacitance (which swamps the self winding capacitance) and find the resonant frequency and work out the inductance from that.

In the past I have done a number of projects where a large inductance was difficult to work out and measure, examples being ignition coil secondaries and tube audio inter-stage transformers, where two significant inductances and one with significant self winding capacitances are on the same core and influence each other. Much to my astonishment, the equation for the audio band-pass characteristic of a tube inter-stage transformer (like those used in 1920's tube radios) was not readily available, so I had to work it out myself. (The transformer models used in the audiophile industry are not suited to a transformer with an unloaded secondary). If you are interested, the equation is included here on page 11 to 17 of this vintage radio article:

http://worldphaco.com/uploads/THE_GREBE_MU-1.pdf

Another trick that can be used to find an inductance value is simply to set up an output stage, typically mosfet circuit to switch a square wave of +/-10V across the inductor and monitor the current by looking at the voltage across a small series resistor and at a frequency that the current doesn't climb too high. The rate of change of current with time, you see and can measure initially with a scope after switching, is equal to the known applied voltage divided by the inductance value.

mpegjohn · 10th Dec 2016, 7:26 pm

Thank you Agnus25, I will try to get my head around that.

So I have compared an Agilent LCR meter reading against my bridge, and they both tally. This was with no DC current.

So taking your advice, I could put a known DC current through, and resonate the choke. Work out its inductance, and then try on my bridge. Is that correct?

John

Argus25 · 11th Dec 2016, 1:47 am

1) "The question in my mind is, is my current source messing with my measurement results".

2) "So taking your advice, I could put a known DC current through, and resonate the choke. Work out its inductance, and then try on my bridge. Is that correct?"

With regards to the two questions above:

1) In many cases its normal that a DC current will alter the measured inductance, especially of iron cored transformers/chokes as the core of the choke starts to operate on a higher part of the B-H curve, and in the end it is possible to saturate the core (magnetically). At that point the inductance is significantly reduced. This is why many transformers, ie for class A audio work & power supply chokes are butt stacked with an air gap to help prevent this effect. Air cored inductors do not have this issue. So its not messing with the measurement results, its just an expected effect.

2) You could resonate your inductance with a parallel capacitor and couple the AC from the generator into it via a large value series resistor >100k to 1M to prevent damping in series with the ground leg of the generator and put the scope across the resistor and then run a DC current through the inductor from your high Z mosfet current source, to see how the resonant frequency alters, and recalculate the inductance values at different currents.

You may be able to use the mosfet DC current source described in the diyaudioprojects articles to bias the inductor under test using the Agilent LCR meter, but before that was a safe thing to do you would need to look at the schematic of their meter to see if it was permissible to have a DC voltage across the inductor or not, without damaging their meter or upsetting the measurement. If it wasn't ok, and you were winding/making your own inductors, you could add an additional isolated winding to the inductor you have made to magnetize the core with some value of ampere-turns and work out the effect of core magnetization that way while still being able to use the Agilent meter.

Radio Wrangler · 11th Dec 2016, 3:11 am

A cheat I've used is to make two of the inductors to be measured, wre them in series and pass the wanted DC through them from a well-decoupled supply. Connect an LCR meter or bridge onto the junction between the inductors via a blocking capacitor and measure. You get the results for two identical parts in parallel so the true inductance is double the reading. ESR is double the reading. Q is right.

Be careful when shutting down. Disconnect the LCR meter or bridge, then bring down the current gently.

David

trobbins · 12th Dec 2016, 6:49 am

I updated comments in my choke measurement doc. relating to accuracy. The issue with using a rectified AC plus DC signal is that the measurement is not a single frequency, but includes rectifier harmonics plus any mains waveform distortion. That situation could be alleviated somewhat by RC filtering before the measurement signal is passed through the choke. The shunt capacitance at twice mains f is negligible as an error in determining the inductive component.

8th Dec 2016, 11:09 am	#1
mpegjohn Hexode Join Date: Aug 2013 Location: Chichester, West Sussex, UK. Posts: 270	Confirmation of DC constant current calibration in a bridge. Hi. To test a choke I am winding I have built an Owens bridge, with a MOSFET constant current source. To check the accuracy of the bridge, I have compared some chokes, against an Agilent LCR meter we have at work. But this does not have any DC current source. The question in my mind is, is my current source messing with my measurement results. So, I would love to know if anyone has a bridge or LCR meter with a CC source, that they know is good, that will allow me to send my choke, and do a few measurements so I can compare and contrast. It doesn't need to be my choke either, I just need a reference piece, with known inductance at specified DC currents. Any help will be appreciated. The Owens bridge is a series of 4 articles here:- http://diyaudioprojects.com/mirror/m...com/sbench101/ Regards, John.

8th Dec 2016, 11:18 am	#2
trobbins Heptode Join Date: Oct 2009 Location: Melbourne Australia Posts: 901	Re: confirmation of DC constant current calibration in a bridge John, Do you need an accurate measurement of inductance at a given DC current, and what are your nominal levels (eg. mA, or A, mH or H, or ...) ? Or does your application allow some latitude - eg. if you were building a power supply filter choke? Ciao, Tim

8th Dec 2016, 11:27 am	#3
mpegjohn Hexode Join Date: Aug 2013 Location: Chichester, West Sussex, UK. Posts: 270	Re: confirmation of DC constant current calibration in a bridge Tim, I need an accurate reading of inductance at various DC levels. Inductance ~10H DC 0-100mA, for the inductors I have, but if not using mine, then any inductor > say 1H with Idc upto 200mA. The actual inductor I am not interested in, just some form of cross checking my bridge setup. John

8th Dec 2016, 2:53 pm	#4
PJL Dekatron Join Date: Sep 2005 Location: Seaford, East Sussex, UK. Posts: 5,997	Re: Confirmation of DC constant current calibration in a bridge. How about measuring the 'inductance' of a resistor with the same resistance as the inductance you want to test over the current range?

8th Dec 2016, 3:27 pm	#5
mpegjohn Hexode Join Date: Aug 2013 Location: Chichester, West Sussex, UK. Posts: 270	Re: Confirmation of DC constant current calibration in a bridge. Not a bad idea at all.

8th Dec 2016, 7:50 pm	#6
mpegjohn Hexode Join Date: Aug 2013 Location: Chichester, West Sussex, UK. Posts: 270	Re: Confirmation of DC constant current calibration in a bridge. Just had a thought though, if I have a 10H choke at 100Hz that's 6K. at 100mA that will need at least 600V, my CC source has 110V max. Maybe I could use Tim Robins method of choke measurement to get fixed points of DC and AC.

10th Dec 2016, 2:43 pm	#7
Argus25 No Longer a Member Join Date: Oct 2016 Location: Maroochydore, Queensland, Australia. Posts: 2,679	Re: Confirmation of DC constant current calibration in a bridge. As you note, if you could get an accurate assessment of a sample inductance, by another method, then you could check your bridge calibration. Your Owen's bridge will probably be very accurate for small range inductances. Problems can crop up measuring large inductances though on many bridges and meters. There is one way around the problem of large inductors and measuring or estimating their value. I have used this method in the past, it works: If an experimental inductor is wound with much thicker wire and far fewer turns on the same bobbin and core, provided the physical space occupied by the winding on the bobbin is similar to the final high inductance version (usually the most efficient design uses all the available free cross sectional area of the bobbin) you can measure the much lower inductance value very easily and accurately on most inductance meters and bridges. And since the inductance is proportional to the square of the number of turns it is easy then to calculate the required number of turns for the final large inductor value you want. And the inductance should be correct, even if it is an extremely high value. Then you can check it with the bridge to see how accurate the bridge is in the higher ranges. You can also check the effect of DC current on the lower inductance version, being mindful of the ampere-turns effect, which will be the same on the final large inductance so you can transpose the values because you will also know the turns numbers and the ratio of the DC resistances of the two inductor versions and any applied DC voltages.You will need a much lower DC voltage for the smaller value version of the inductor for this test. So you could use that method to check the performance of your bridge for small vs high value inductors. For the low range values I would simply recommend comparing it to a commercial meter. One problem with large inductors that have many turns of fine wire is that the self winding capacitance and coil DC resistance can become significant. The capacitance can foul up the reading on some meters & bridges depending on the operating frequency and principle of the measuring system. One method of measurement is to tune them with a 0.1uF capacitance (which swamps the self winding capacitance) and find the resonant frequency and work out the inductance from that. In the past I have done a number of projects where a large inductance was difficult to work out and measure, examples being ignition coil secondaries and tube audio inter-stage transformers, where two significant inductances and one with significant self winding capacitances are on the same core and influence each other. Much to my astonishment, the equation for the audio band-pass characteristic of a tube inter-stage transformer (like those used in 1920's tube radios) was not readily available, so I had to work it out myself. (The transformer models used in the audiophile industry are not suited to a transformer with an unloaded secondary). If you are interested, the equation is included here on page 11 to 17 of this vintage radio article: http://worldphaco.com/uploads/THE_GREBE_MU-1.pdf Another trick that can be used to find an inductance value is simply to set up an output stage, typically mosfet circuit to switch a square wave of +/-10V across the inductor and monitor the current by looking at the voltage across a small series resistor and at a frequency that the current doesn't climb too high. The rate of change of current with time, you see and can measure initially with a scope after switching, is equal to the known applied voltage divided by the inductance value.

10th Dec 2016, 7:26 pm	#8
mpegjohn Hexode Join Date: Aug 2013 Location: Chichester, West Sussex, UK. Posts: 270	Re: Confirmation of DC constant current calibration in a bridge. Thank you Agnus25, I will try to get my head around that. So I have compared an Agilent LCR meter reading against my bridge, and they both tally. This was with no DC current. So taking your advice, I could put a known DC current through, and resonate the choke. Work out its inductance, and then try on my bridge. Is that correct? John

11th Dec 2016, 1:47 am	#9
Argus25 No Longer a Member Join Date: Oct 2016 Location: Maroochydore, Queensland, Australia. Posts: 2,679	Re: Confirmation of DC constant current calibration in a bridge. 1) "The question in my mind is, is my current source messing with my measurement results". 2) "So taking your advice, I could put a known DC current through, and resonate the choke. Work out its inductance, and then try on my bridge. Is that correct?" With regards to the two questions above: 1) In many cases its normal that a DC current will alter the measured inductance, especially of iron cored transformers/chokes as the core of the choke starts to operate on a higher part of the B-H curve, and in the end it is possible to saturate the core (magnetically). At that point the inductance is significantly reduced. This is why many transformers, ie for class A audio work & power supply chokes are butt stacked with an air gap to help prevent this effect. Air cored inductors do not have this issue. So its not messing with the measurement results, its just an expected effect. 2) You could resonate your inductance with a parallel capacitor and couple the AC from the generator into it via a large value series resistor >100k to 1M to prevent damping in series with the ground leg of the generator and put the scope across the resistor and then run a DC current through the inductor from your high Z mosfet current source, to see how the resonant frequency alters, and recalculate the inductance values at different currents. You may be able to use the mosfet DC current source described in the diyaudioprojects articles to bias the inductor under test using the Agilent LCR meter, but before that was a safe thing to do you would need to look at the schematic of their meter to see if it was permissible to have a DC voltage across the inductor or not, without damaging their meter or upsetting the measurement. If it wasn't ok, and you were winding/making your own inductors, you could add an additional isolated winding to the inductor you have made to magnetize the core with some value of ampere-turns and work out the effect of core magnetization that way while still being able to use the Agilent meter.

11th Dec 2016, 3:11 am	#10
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,902	Re: Confirmation of DC constant current calibration in a bridge. A cheat I've used is to make two of the inductors to be measured, wre them in series and pass the wanted DC through them from a well-decoupled supply. Connect an LCR meter or bridge onto the junction between the inductors via a blocking capacitor and measure. You get the results for two identical parts in parallel so the true inductance is double the reading. ESR is double the reading. Q is right. Be careful when shutting down. Disconnect the LCR meter or bridge, then bring down the current gently. David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

12th Dec 2016, 6:49 am	#11
trobbins Heptode Join Date: Oct 2009 Location: Melbourne Australia Posts: 901	Re: Confirmation of DC constant current calibration in a bridge. I updated comments in my choke measurement doc. relating to accuracy. The issue with using a rectified AC plus DC signal is that the measurement is not a single frequency, but includes rectifier harmonics plus any mains waveform distortion. That situation could be alleviated somewhat by RC filtering before the measurement signal is passed through the choke. The shunt capacitance at twice mains f is negligible as an error in determining the inductive component.