I used to know all this many years ago... how to work out the flux
density, and choose a material which doesn't saturate, but I have long
forgotten!

It needs to be good with a 13V peak square wave, 500Hz.

So we have 13V across 8mH for 1ms, which from v = L (di/dt) yields
1.625A. A quick hack in LTspice confirms this, for the first 5 cycles.
I had a lot of trouble generating a square wave from -13V to +13V :)
so I am not sure what the current waveform will look like when you
switch to -13V when the current flowing is still 1.625A; I suspect it
will not ever exceed 1.625A later though. With a 13V peak sinewave (a
predefined function in LTspice) it looked to be just under 1A, but all
positive which is obviously BS.

The bit I have forgotten is how to calculate the flux density in the
core. I would prefer the whole thing to be something the size of an
RM10 core
https://www.farnell.com/datasheets/528671.pdf
In 3C95 material you have 5500nH/T^2.

A toroid would have less stray flux, and again there is a vast choice.
I have a bag of these from many years ago
https://www.tme.eu/gb/details/tn16_9.6-3f3/ring-ferrites/ferroxcube/tn16-9-6-6-3-3f3/
which with 83 turns would achieve 8mH.

What I don't get is how much current this will carry. 83 turns at 1.6A
is 132 AT which sounds an awful lot. I thus suspect I will need a
bigger core, probably iron.

Looking on Ebay for ready made stuff, 10mH, I see e.g.
https://www.ebay.co.uk/itm/115852064002
but I am instinctively damn sure that won't carry the current.

Another option is some toroidal transformer, and ignore the existing
winding and put more turns on it. Ebay is full of vintage transformers
but most are pretty big.

I was going to wind the TN16 with some turns and see what it does. I
have an HP 3314 pulse generator and a power amp which can output 9V
peak.

The 8mH needs to be +/- 0.3mH. It was determined experimentally using
this amazing thing
https://www.ebay.co.uk/itm/224101042113
but that obviously contains massive magnetics - it weighs about 1kg.

I also have an LCR meter.

I'd be grateful for any suggestions.

On Tue, 30 Jan 2024 16:13:41 +0000, Peter
<occassionally-confused@nospam.co.uk> wrote:

>I used to know all this many years ago... how to work out the flux
>density, and choose a material which doesn't saturate, but I have long
>forgotten!
>
>It needs to be good with a 13V peak square wave, 500Hz.
>
>So we have 13V across 8mH for 1ms, which from v = L (di/dt) yields
>1.625A. A quick hack in LTspice confirms this, for the first 5 cycles.
>I had a lot of trouble generating a square wave from -13V to +13V :)
>so I am not sure what the current waveform will look like when you
>switch to -13V when the current flowing is still 1.625A; I suspect it
>will not ever exceed 1.625A later though. With a 13V peak sinewave (a
>predefined function in LTspice) it looked to be just under 1A, but all
>positive which is obviously BS.
>
>The bit I have forgotten is how to calculate the flux density in the
>core. I would prefer the whole thing to be something the size of an
>RM10 core
>https://www.farnell.com/datasheets/528671.pdf
>In 3C95 material you have 5500nH/T^2.
>
>A toroid would have less stray flux, and again there is a vast choice.
>I have a bag of these from many years ago
>https://www.tme.eu/gb/details/tn16_9.6-3f3/ring-ferrites/ferroxcube/tn16-9-6-6-3-3f3/
>which with 83 turns would achieve 8mH.
>
>What I don't get is how much current this will carry. 83 turns at 1.6A
>is 132 AT which sounds an awful lot. I thus suspect I will need a
>bigger core, probably iron.
>
>Looking on Ebay for ready made stuff, 10mH, I see e.g.
>https://www.ebay.co.uk/itm/115852064002
>but I am instinctively damn sure that won't carry the current.
>
>Another option is some toroidal transformer, and ignore the existing
>winding and put more turns on it. Ebay is full of vintage transformers
>but most are pretty big.
>
>I was going to wind the TN16 with some turns and see what it does. I
>have an HP 3314 pulse generator and a power amp which can output 9V
>peak.
>
>The 8mH needs to be +/- 0.3mH. It was determined experimentally using
>this amazing thing
>https://www.ebay.co.uk/itm/224101042113
>but that obviously contains massive magnetics - it weighs about 1kg.
>
>I also have an LCR meter.
>
>I'd be grateful for any suggestions.

To make a square wave in LT Spice, use a voltage source, right-click
'advanced' and then 'pulse' and set the params to get what looks
right.

Or run a sine wave through a BV block with some equation. That makes
it easy to change the frequency and keep 50% duty cycle.

The easy way to 'design' an inductor is to buy one. Most suppliers
specify L and R and max current and saturation current. Even if you
enjoy winding inductors, a standard part is a good sanity check.

LCR meters usually lie when measuring power inductors.

What's your application?

John Larkin <jl@997PotHill.com> wrote

>To make a square wave in LT Spice, use a voltage source, right-click
>'advanced' and then 'pulse' and set the params to get what looks
>right.
>
>Or run a sine wave through a BV block with some equation. That makes
>it easy to change the frequency and keep 50% duty cycle.
>
>The easy way to 'design' an inductor is to buy one. Most suppliers
>specify L and R and max current and saturation current. Even if you
>enjoy winding inductors, a standard part is a good sanity check.
>
>LCR meters usually lie when measuring power inductors.
>
>What's your application?

It is to reduce the output from an LVDT, by putting this in series
with the input. The reason for reducing the output is a bit obscure...

On Tue, 30 Jan 2024 20:32:25 +0000, Peter
<occassionally-confused@nospam.co.uk> wrote:

>
>John Larkin <jl@997PotHill.com> wrote
>
>>To make a square wave in LT Spice, use a voltage source, right-click
>>'advanced' and then 'pulse' and set the params to get what looks
>>right.
>>
>>Or run a sine wave through a BV block with some equation. That makes
>>it easy to change the frequency and keep 50% duty cycle.
>>
>>The easy way to 'design' an inductor is to buy one. Most suppliers
>>specify L and R and max current and saturation current. Even if you
>>enjoy winding inductors, a standard part is a good sanity check.
>>
>>LCR meters usually lie when measuring power inductors.
>>
>>What's your application?
>
>It is to reduce the output from an LVDT, by putting this in series
>with the input. The reason for reducing the output is a bit obscure...

We do a lot of LVDT measurement and simulation, usually at higher
frequency but occasionally at 400 Hz.

You could use a step-down transformer instead of a series inductor,
for less phase shift and better linearity and a predictable ratio.
Some little audio transformer.

Ir maybe a resistive voltage divider.

On Wednesday, January 31, 2024 at 7:44:15 AM UTC+11, john larkin wrote:
> On Tue, 30 Jan 2024 20:32:25 +0000, Peter
> <occassional...@nospam.co.uk> wrote:
>
> >
> >John Larkin <j...@997PotHill.com> wrote
> >
> >>To make a square wave in LT Spice, use a voltage source, right-click
> >>'advanced' and then 'pulse' and set the params to get what looks
> >>right.
> >>
> >>Or run a sine wave through a BV block with some equation. That makes
> >>it easy to change the frequency and keep 50% duty cycle.
> >>
> >>The easy way to 'design' an inductor is to buy one. Most suppliers
> >>specify L and R and max current and saturation current. Even if you
> >>enjoy winding inductors, a standard part is a good sanity check.
> >>
> >>LCR meters usually lie when measuring power inductors.
> >>
> >>What's your application?
> >
> >It is to reduce the output from an LVDT, by putting this in series
> >with the input. The reason for reducing the output is a bit obscure...

John Larkin is right about one thing - the easiest way of reducing the output from an LVDT is with transformer.

Using it to scale down the drive voltage to the LVDT probably makes more sense than using it to scale down the output, if only because the LVDT would run cooler.

An RM10 has a magnetic path length of 44mm

https://www.tdk-electronics.tdk.com/inf/80/db/fer/rm_10.pdf

Your 132A.turns is thus 3000 A.T/metre which is a lot - you don't really want to go above 50A.T/metre.

You might be able save the idea with a huge gap and lots more turns but I doubt it. I wouldn't bother trying even for the most pointy-headed boss.

https://www.tdk-electronics.tdk.com/download/519704/069c210d0363d7b4682d9ff22c2ba503/ferrites-and-accessories-db-130501.pdf

gives all the information.

--
Bill Sloman, Sydney

On Tue, 30 Jan 2024 16:13:41 +0000, Peter
<occassionally-confused@nospam.co.uk> wrote:

>I used to know all this many years ago... how to work out the flux
>density, and choose a material which doesn't saturate, but I have long
>forgotten!

Don't be surprised. It's common to have to brush up on basics, if
you don't use them regularly.
>
>It needs to be good with a 13V peak square wave, 500Hz.
>
>So we have 13V across 8mH for 1ms, which from v = L (di/dt) yields
>1.625A. A quick hack in LTspice confirms this, for the first 5 cycles.
>I had a lot of trouble generating a square wave from -13V to +13V :)
>so I am not sure what the current waveform will look like when you
>switch to -13V when the current flowing is still 1.625A; I suspect it
>will not ever exceed 1.625A later though. With a 13V peak sinewave (a
>predefined function in LTspice) it looked to be just under 1A, but all
>positive which is obviously BS.

Your AC source is seen by the sim to have the polarity of the first
+ pulse. The -pulse that follows can only return mag current to
zero, hence a net +DC can show up in early portions of the sim.
Ad a load to get AC-only current, eventually.
>
>The bit I have forgotten is how to calculate the flux density in the
>core. I would prefer the whole thing to be something the size of an
>RM10 core
>https://www.farnell.com/datasheets/528671.pdf
>In 3C95 material you have 5500nH/T^2.
>
>A toroid would have less stray flux, and again there is a vast choice.
>I have a bag of these from many years ago
>https://www.tme.eu/gb/details/tn16_9.6-3f3/ring-ferrites/ferroxcube/tn16-9-6-6-3-3f3/
>which with 83 turns would achieve 8mH.
>
>What I don't get is how much current this will carry. 83 turns at 1.6A
>is 132 AT which sounds an awful lot. I thus suspect I will need a
>bigger core, probably iron.

RM10 DC saturation current will depend on turns count and
gap. Saturation flux in ferrite is ~0.33T.

B = u N I / lm

B = flux density in Teslas
u = 4.pi.E-7
N = turns count
I = current in Amps
lm = magnetic path length (or gap width, if present) in meters.

For pulsed DC

deltsB = V T / N Ae

deltaB = flux density change in Teslas
V = applied volts in Volts
T = time in seconds
N = turns count
Ae = cross-sectional area of flux path im meters^2

inductance of gapped structure

L = 4.pi.E-7 N^2 Ae / lg

L = inductance in Henries
N = turns count
Ae = cross-sectional area of flux path im meters^2
lg = gap width in meters

>
>Looking on Ebay for ready made stuff, 10mH, I see e.g.
>https://www.ebay.co.uk/itm/115852064002
>but I am instinctively damn sure that won't carry the current.
>
>Another option is some toroidal transformer, and ignore the existing
>winding and put more turns on it. Ebay is full of vintage transformers
>but most are pretty big.

Solutions will depend on whether or not you intend to use the part
in AC-only, small signal situations, or whether the part is intended
to do some work in a resonant power or filter circuit.
>
>I was going to wind the TN16 with some turns and see what it does. I
>have an HP 3314 pulse generator and a power amp which can output 9V
>peak.
>
>The 8mH needs to be +/- 0.3mH. It was determined experimentally using
>this amazing thing
>https://www.ebay.co.uk/itm/224101042113
>but that obviously contains massive magnetics - it weighs about 1kg.
>
>I also have an LCR meter.
>
>I'd be grateful for any suggestions.

Check what's on offer in distributors' catalogs. This will give
you a ball park idea of the size and shape of stuff you'll be
aiming for and may offer a simple solution off the shelf.

RL

On Wed, 31 Jan 2024 11:30:01 -0500, legg <legg@nospam.magma.ca> wrote:

>On Tue, 30 Jan 2024 16:13:41 +0000, Peter
><occassionally-confused@nospam.co.uk> wrote:
>
>>I used to know all this many years ago... how to work out the flux
>>density, and choose a material which doesn't saturate, but I have long
>>forgotten!
>
>Don't be surprised. It's common to have to brush up on basics, if
>you don't use them regularly.
>>
>>It needs to be good with a 13V peak square wave, 500Hz.
>>
>>So we have 13V across 8mH for 1ms, which from v = L (di/dt) yields
>>1.625A. A quick hack in LTspice confirms this, for the first 5 cycles.
>>I had a lot of trouble generating a square wave from -13V to +13V :)
>>so I am not sure what the current waveform will look like when you
>>switch to -13V when the current flowing is still 1.625A; I suspect it
>>will not ever exceed 1.625A later though. With a 13V peak sinewave (a
>>predefined function in LTspice) it looked to be just under 1A, but all
>>positive which is obviously BS.
>
>Your AC source is seen by the sim to have the polarity of the first
>+ pulse. The -pulse that follows can only return mag current to
>zero, hence a net +DC can show up in early portions of the sim.
>Ad a load to get AC-only current, eventually.
>>
>>The bit I have forgotten is how to calculate the flux density in the
>>core. I would prefer the whole thing to be something the size of an
>>RM10 core
>>https://www.farnell.com/datasheets/528671.pdf
>>In 3C95 material you have 5500nH/T^2.
>>
>>A toroid would have less stray flux, and again there is a vast choice.
>>I have a bag of these from many years ago
>>https://www.tme.eu/gb/details/tn16_9.6-3f3/ring-ferrites/ferroxcube/tn16-9-6-6-3-3f3/
>>which with 83 turns would achieve 8mH.
>>
>>What I don't get is how much current this will carry. 83 turns at 1.6A
>>is 132 AT which sounds an awful lot. I thus suspect I will need a
>>bigger core, probably iron.
>
>RM10 DC saturation current will depend on turns count and
>gap. Saturation flux in ferrite is ~0.33T.
>
>B = u N I / lm
>
>B = flux density in Teslas
>u = 4.pi.E-7
>N = turns count
>I = current in Amps
>lm = magnetic path length (or gap width, if present) in meters.
>
>
>For pulsed DC
>
>deltsB = V T / N Ae
>
>deltaB = flux density change in Teslas
>V = applied volts in Volts
>T = time in seconds
>N = turns count
>Ae = cross-sectional area of flux path im meters^2
>
>
>inductance of gapped structure
>
>L = 4.pi.E-7 N^2 Ae / lg
>
>L = inductance in Henries
>N = turns count
>Ae = cross-sectional area of flux path im meters^2
>lg = gap width in meters
>
>
>>
>>Looking on Ebay for ready made stuff, 10mH, I see e.g.
>>https://www.ebay.co.uk/itm/115852064002
>>but I am instinctively damn sure that won't carry the current.
>>
>>Another option is some toroidal transformer, and ignore the existing
>>winding and put more turns on it. Ebay is full of vintage transformers
>>but most are pretty big.
>
>Solutions will depend on whether or not you intend to use the part
>in AC-only, small signal situations, or whether the part is intended
>to do some work in a resonant power or filter circuit.
>>
>>I was going to wind the TN16 with some turns and see what it does. I
>>have an HP 3314 pulse generator and a power amp which can output 9V
>>peak.
>>
>>The 8mH needs to be +/- 0.3mH. It was determined experimentally using
>>this amazing thing
>>https://www.ebay.co.uk/itm/224101042113
>>but that obviously contains massive magnetics - it weighs about 1kg.
>>
>>I also have an LCR meter.
>>
>>I'd be grateful for any suggestions.
>
>Check what's on offer in distributors' catalogs. This will give
>you a ball park idea of the size and shape of stuff you'll be
>aiming for and may offer a simple solution off the shelf.
>
>RL

It is unusual to run a square wave into an LVDT. If the source is a
too-big square wave, a simple circuit would shape it into the
appropriate amplitude sine. Two or three parts.

On Thursday, February 1, 2024 at 4:05:45 AM UTC+11, John Larkin wrote:
> On Wed, 31 Jan 2024 11:30:01 -0500, legg <le...@nospam.magma.ca> wrote:
> >On Tue, 30 Jan 2024 16:13:41 +0000, Peter<occassional...@nospam.co.uk> wrote:

<snip>

> It is unusual to run a square wave into an LVDT. If the source is a too-big square wave, a simple circuit would shape it into the
appropriate amplitude sine. Two or three parts.

It's not so much unusual as downright bonkers.

When I was a young and non-too-well-informed graduate student I excited an LVDT with a triangular wave (which contains the same odd harmonics, but at least their amplitude drops off faster with frequency). The next time I had to excite an LVDT, some ten years later, I used a Baxandall class D oscillator to make a slightly dirty sine wave (about 5% harmonic content) and the time after that - another ten years later - I invented my current mirror variant of the Baxandall class D oscillator to produce an order of magnitude cleaner sine wave.

http://sophia-electronica.com/BillsBaxandall.html

--
Bill Sloman, Sydney

On Tue, 30 Jan 2024 16:13:41 +0000, Peter
<occassionally-confused@nospam.co.uk> wrote:

>I used to know all this many years ago... how to work out the flux
>density, and choose a material which doesn't saturate, but I have long
>forgotten!
>
>It needs to be good with a 13V peak square wave, 500Hz.
>
<snip>
>core. I would prefer the whole thing to be something the size of an
>RM10 core
>https://www.farnell.com/datasheets/528671.pdf
>In 3C95 material you have 5500nH/T^2.
>
<snip>
>I have a bag of these from many years ago
>https://www.tme.eu/gb/details/tn16_9.6-3f3/ring-ferrites/ferroxcube/tn16-9-6-6-3-3f3/
>which with 83 turns would achieve 8mH.
>
>What I don't get is how much current this will carry.
<snip>
>
>Looking on Ebay for ready made stuff, 10mH, I see e.g.
>https://www.ebay.co.uk/itm/115852064002
>but I am instinctively damn sure that won't carry the current.
>
<snip>
>I was going to wind the TN16 with some turns and see what it does. I
>have an HP 3314 pulse generator and a power amp which can output 9V
>peak.
<snip..

It would be easier to help if you stated the application and known
limits/requirements of size, weight, mounting method, environment etc.

'carry the current' really doesn't tell us anything.

If there are 'found' materials you'd prefer to use . . . and results
from your attempts to do so.

etc.

RL

On Wed, 31 Jan 2024 09:04:20 -0800, John Larkin <jl@997PotHill.com>
wrote:

<snip>
>
>It is unusual to run a square wave into an LVDT. If the source is a
>too-big square wave, a simple circuit would shape it into the
>appropriate amplitude sine. Two or three parts.

I see no reference to an LVDT application in the Peter's posting.

RL

legg <legg@nospam.magma.ca> wrote:

>>It is unusual to run a square wave into an LVDT. If the source is a
>>too-big square wave, a simple circuit would shape it into the
>>appropriate amplitude sine. Two or three parts.
>
>I see no reference to an LVDT application in the Peter's posting.

It is an instrument which contains an LVDT bent into a half circle, to
return an angle.

It is excited with 500Hz, and it is pretty well a square wave. I
didn't design that bit :) And the output is too high, for reasons too
complex to explain. So I need to reduce the output somehow.

On Thu, 01 Feb 2024 09:35:56 -0500, legg <legg@nospam.magma.ca> wrote:

>On Wed, 31 Jan 2024 09:04:20 -0800, John Larkin <jl@997PotHill.com>
>wrote:
>
>
><snip>
>>
>>It is unusual to run a square wave into an LVDT. If the source is a
>>too-big square wave, a simple circuit would shape it into the
>>appropriate amplitude sine. Two or three parts.
>
>I see no reference to an LVDT application in the Peter's posting.
>
>RL

Subject: Re: How to make an 8mH inductor which can handle 13V peak
square wave without saturation
Date: Tue, 30 Jan 2024 20:32:25 +0000

On Thu, 01 Feb 2024 08:01:11 -0800, John Larkin <jl@997PotHill.com>
wrote:

>On Thu, 01 Feb 2024 09:35:56 -0500, legg <legg@nospam.magma.ca> wrote:
>
>>On Wed, 31 Jan 2024 09:04:20 -0800, John Larkin <jl@997PotHill.com>
>>wrote:
>>
>>
>><snip>
>>>
>>>It is unusual to run a square wave into an LVDT. If the source is a
>>>too-big square wave, a simple circuit would shape it into the
>>>appropriate amplitude sine. Two or three parts.
>>
>>I see no reference to an LVDT application in the Peter's posting.
>>
>>RL
>
>Subject: Re: How to make an 8mH inductor which can handle 13V peak
>square wave without saturation
>Date: Tue, 30 Jan 2024 20:32:25 +0000

Not on this reader. I have only:

Tue, 30 Jan 2024 16:13:41

RL

On Thu, 01 Feb 2024 15:37:48 +0000, Peter
<occassionally-confused@nospam.co.uk> wrote:

>
> legg <legg@nospam.magma.ca> wrote:
>
>>>It is unusual to run a square wave into an LVDT. If the source is a
>>>too-big square wave, a simple circuit would shape it into the
>>>appropriate amplitude sine. Two or three parts.
>>
>>I see no reference to an LVDT application in the Peter's posting.
>
>It is an instrument which contains an LVDT bent into a half circle, to
>return an angle.
>
>It is excited with 500Hz, and it is pretty well a square wave. I
>didn't design that bit :) And the output is too high, for reasons too
>complex to explain. So I need to reduce the output somehow.

So you've got 500Hz @ 13V (pk?). and you can't adjust the
source (why not?)

AC or DC?

You could reduce the output phase width with a saturable
reactor.

For pulsating DC, it might be a regulated 'set' reactor.

For AC, it could be a bit more complicated, unless
you were satisfied with just a fixed PW blocker.
That might drift a bit with load or temperature.
Gets physically bigger the more voltseconds you want
to block, for the same current.

The 'complicated' saturable reactor for AC begins to
look like a mag amp.

RL

On Thu, 01 Feb 2024 17:07:04 -0500, legg <legg@nospam.magma.ca> wrote:

>On Thu, 01 Feb 2024 08:01:11 -0800, John Larkin <jl@997PotHill.com>
>wrote:
>
>>On Thu, 01 Feb 2024 09:35:56 -0500, legg <legg@nospam.magma.ca> wrote:
>>
>>>On Wed, 31 Jan 2024 09:04:20 -0800, John Larkin <jl@997PotHill.com>
>>>wrote:
>>>
>>>
>>><snip>
>>>>
>>>>It is unusual to run a square wave into an LVDT. If the source is a
>>>>too-big square wave, a simple circuit would shape it into the
>>>>appropriate amplitude sine. Two or three parts.
>>>
>>>I see no reference to an LVDT application in the Peter's posting.
>>>
>>>RL
>>
>>Subject: Re: How to make an 8mH inductor which can handle 13V peak
>>square wave without saturation
>>Date: Tue, 30 Jan 2024 20:32:25 +0000
>
>Not on this reader. I have only:
>
>Tue, 30 Jan 2024 16:13:41
>
>RL

Anyhow, he said he wanted to reduce the drive level into an LVDT.

I think he could use a simple LC or RLC to make a nice lower-amplitude
sine from his big square wave.

On 2/02/2024 2:37 am, Peter wrote:
>
> legg <legg@nospam.magma.ca> wrote:
>
>>> It is unusual to run a square wave into an LVDT. If the source is a
>>> too-big square wave, a simple circuit would shape it into the
>>> appropriate amplitude sine. Two or three parts.
>>
>> I see no reference to an LVDT application in the Peter's posting.
>
> It is an instrument which contains an LVDT bent into a half circle, to
> return an angle.
>
> It is excited with 500Hz, and it is pretty well a square wave. I
> didn't design that bit :) And the output is too high, for reasons too
> complex to explain. So I need to reduce the output somehow.

It would make more sense to excite it with a lower amplitude sine wave.
The angle sensing differential transformer will run cooler - those
higher harmonics generate extra heat in the core material.

It's a bigger change than your boss had in mind, but would give you an
appreciably better system.

--
Bill Sloman, Sydney

On Thu, 01 Feb 2024 15:37:48 +0000, Peter
<occassionally-confused@nospam.co.uk> wrote:

>
> legg <legg@nospam.magma.ca> wrote:
>
>>>It is unusual to run a square wave into an LVDT. If the source is a
>>>too-big square wave, a simple circuit would shape it into the
>>>appropriate amplitude sine. Two or three parts.
>>
>>I see no reference to an LVDT application in the Peter's posting.
>
>It is an instrument which contains an LVDT bent into a half circle, to
>return an angle.
>
>It is excited with 500Hz, and it is pretty well a square wave. I
>didn't design that bit :) And the output is too high, for reasons too
>complex to explain. So I need to reduce the output somehow.

About LVDTs, I know nada.

RL

On Fri, 02 Feb 2024 10:08:46 -0500, legg <legg@nospam.magma.ca> wrote:

>On Thu, 01 Feb 2024 15:37:48 +0000, Peter
><occassionally-confused@nospam.co.uk> wrote:
>
>>
>> legg <legg@nospam.magma.ca> wrote:
>>
>>>>It is unusual to run a square wave into an LVDT. If the source is a
>>>>too-big square wave, a simple circuit would shape it into the
>>>>appropriate amplitude sine. Two or three parts.
>>>
>>>I see no reference to an LVDT application in the Peter's posting.
>>
>>It is an instrument which contains an LVDT bent into a half circle, to
>>return an angle.
>>
>>It is excited with 500Hz, and it is pretty well a square wave. I
>>didn't design that bit :) And the output is too high, for reasons too
>>complex to explain. So I need to reduce the output somehow.
>
>About LVDTs, I know nada.
>
>RL

They can be interesting.

https://en.wikipedia.org/wiki/Linear_variable_differential_transformer

but there are many varieties and most are not well documented. I get
them and test them to understand them better. I recently bought some
from ebay, the kind that are used in machine tools, with micro-inch
resolution.

I'm planning another LVDT/synchro measurement/simulation box and maybe
machine tools could be a new market. For some reason the LVDTs and
synchros used in machining run at much higher frequencies than the
aerospace stuff, maybe because the cables are typically shorter.

There are LVDT-based inclinometers too.

On 30/01/2024 16:13, Peter wrote:

<snip>

> A toroid would have less stray flux, and again there is a vast choice.
> I have a bag of these from many years ago
> https://www.tme.eu/gb/details/tn16_9.6-3f3/ring-ferrites/ferroxcube/tn16-9-6-6-3-3f3/
> which with 83 turns would achieve 8mH.

....but nowhere near enough current. Ferroxcube have (or had) a free
downloadable calculator.

As you have already some cores, you could stack several into a tube
shape. I've done this where the available space made it the only game
in town. Tricky to wind but fun, FSVO 'fun'.

--
Cheers
Clive

On 01/02/2024 23:17, john larkin wrote:

<snip>

> Anyhow, he said he wanted to reduce the drive level into an LVDT.
>
> I think he could use a simple LC or RLC to make a nice lower-amplitude
> sine from his big square wave.
>
Depends how the demodulation works, it's often synched to the drive
signal and there may be no phase adjustment.

--
Cheers
Clive

On Saturday, February 3, 2024 at 3:02:00 AM UTC+11, John Larkin wrote:
> On Fri, 02 Feb 2024 10:08:46 -0500, legg <le...@nospam.magma.ca> wrote:
> >On Thu, 01 Feb 2024 15:37:48 +0000, Peter <occassional...@nospam.co.uk> wrote:
> >> legg <le...@nospam.magma.ca> wrote:

<snip>

> They can be interesting.
>
> https://en.wikipedia.org/wiki/Linear_variable_differential_transformer

It's not a great write up. It talks about the high permeability moving core which changes the coupling between the primary and the two opposing secondaries. then goes on to say that they can be used at high temperatures when the moving core isn't necessarily going to be ferromagnetic any longer

https://en.wikipedia.org/wiki/Curie_temperature

> but there are many varieties and most are not well documented. I get
> them and test them to understand them better. I recently bought some
> from ebay, the kind that are used in machine tools, with micro-inch
> resolution.

They don't have any built-in resolution. They provide a strictly analog representation of displacement.

The machine tools at Cambridge Instruments mostly used Heidenhain optical Moire-fringe displacement transducers - the foreman had a Sony displacement transducer on his lathe which kept track of the alternation magnetic field impressed on a length of hard ferromagnetic wire, which didn't get messed up by splashes of cutting fluid.

We used that to monitor where the moving stage of our electron beam tester was - the software guys liked its data interface better.

> I'm planning another LVDT/synchro measurement/simulation box and maybe
> machine tools could be a new market. For some reason the LVDTs and
> synchros used in machining run at much higher frequencies than the
> aerospace stuff, maybe because the cables are typically shorter.
>
> There are LVDT-based inclinometers too.

Aerospace is fixated on 400Hz - it's their 50/60Hz household A/C.

If you used a ferrite moving core with a fairly high DC resistance you can use higher excitation frequencies without heating the moving core and making it expand.

The LVDT we used in our GaAs crystal puller used to move a bit as it warmed up, until I dropped the excitation amplitude when I reworked the electronics that excited it and extracted the signal.

--
Bill Sloman, Sydney

On Fri, 2 Feb 2024 16:20:30 +0000, Clive Arthur
<clive@nowaytoday.co.uk> wrote:

>On 01/02/2024 23:17, john larkin wrote:
>
><snip>
>
>> Anyhow, he said he wanted to reduce the drive level into an LVDT.
>>
>> I think he could use a simple LC or RLC to make a nice lower-amplitude
>> sine from his big square wave.
>>
>Depends how the demodulation works, it's often synched to the drive
>signal and there may be no phase adjustment.

An RLC can make zero phase shift, or lead or lag. The user would have
to consider that. Maybe the original system assumes square waves!

The problem is underspecified.

On Fri, 2 Feb 2024 16:17:45 +0000, Clive Arthur
<clive@nowaytoday.co.uk> wrote:

>On 30/01/2024 16:13, Peter wrote:
>
><snip>
>
>> A toroid would have less stray flux, and again there is a vast choice.
>> I have a bag of these from many years ago
>> https://www.tme.eu/gb/details/tn16_9.6-3f3/ring-ferrites/ferroxcube/tn16-9-6-6-3-3f3/
>> which with 83 turns would achieve 8mH.
>
>...but nowhere near enough current. Ferroxcube have (or had) a free
>downloadable calculator.
>
>As you have already some cores, you could stack several into a tube
>shape. I've done this where the available space made it the only game
>in town. Tricky to wind but fun, FSVO 'fun'.

An LVDT is not likely to need a lot of excitation current.

On Fri, 02 Feb 2024 08:00:38 -0800, John Larkin <jl@997PotHill.com>
wrote:

>On Fri, 02 Feb 2024 10:08:46 -0500, legg <legg@nospam.magma.ca> wrote:
>
>>On Thu, 01 Feb 2024 15:37:48 +0000, Peter
>><occassionally-confused@nospam.co.uk> wrote:
>>
>>>
>>> legg <legg@nospam.magma.ca> wrote:
>>>
>>>>>It is unusual to run a square wave into an LVDT. If the source is a
>>>>>too-big square wave, a simple circuit would shape it into the
>>>>>appropriate amplitude sine. Two or three parts.
>>>>
>>>>I see no reference to an LVDT application in the Peter's posting.
>>>
>>>It is an instrument which contains an LVDT bent into a half circle, to
>>>return an angle.
>>>
>>>It is excited with 500Hz, and it is pretty well a square wave. I
>>>didn't design that bit :) And the output is too high, for reasons too
>>>complex to explain. So I need to reduce the output somehow.
>>
>>About LVDTs, I know nada.
>>
>>RL
>
>They can be interesting.
>
>https://en.wikipedia.org/wiki/Linear_variable_differential_transformer
>
>but there are many varieties and most are not well documented. I get
>them and test them to understand them better. I recently bought some
>from ebay, the kind that are used in machine tools, with micro-inch
>resolution.
>
>I'm planning another LVDT/synchro measurement/simulation box and maybe
>machine tools could be a new market. For some reason the LVDTs and
>synchros used in machining run at much higher frequencies than the
>aerospace stuff, maybe because the cables are typically shorter.
>
>There are LVDT-based inclinometers too.

Yeah, there are a host of things I could fiddle with, but I like
to keep my distractions at least vaguely on track.

RL

On Tue, 30 Jan 2024 16:13:41 +0000, Peter
<occassionally-confused@nospam.co.uk> wrote:

>I used to know all this many years ago... how to work out the flux
>density, and choose a material which doesn't saturate, but I have long
>forgotten!
>
>It needs to be good with a 13V peak square wave, 500Hz.
>
>So we have 13V across 8mH for 1ms, which from v = L (di/dt) yields
>1.625A. A quick hack in LTspice confirms this, for the first 5 cycles.
>I had a lot of trouble generating a square wave from -13V to +13V :)
>so I am not sure what the current waveform will look like when you
>switch to -13V when the current flowing is still 1.625A; I suspect it
>will not ever exceed 1.625A later though. With a 13V peak sinewave (a
>predefined function in LTspice) it looked to be just under 1A, but all
>positive which is obviously BS.
>
>The bit I have forgotten is how to calculate the flux density in the
>core. I would prefer the whole thing to be something the size of an
>RM10 core
>https://www.farnell.com/datasheets/528671.pdf
>In 3C95 material you have 5500nH/T^2.
>
>A toroid would have less stray flux, and again there is a vast choice.
>I have a bag of these from many years ago
>https://www.tme.eu/gb/details/tn16_9.6-3f3/ring-ferrites/ferroxcube/tn16-9-6-6-3-3f3/
>which with 83 turns would achieve 8mH.
>
>What I don't get is how much current this will carry. 83 turns at 1.6A
>is 132 AT which sounds an awful lot. I thus suspect I will need a
>bigger core, probably iron.
>
>Looking on Ebay for ready made stuff, 10mH, I see e.g.
>https://www.ebay.co.uk/itm/115852064002
>but I am instinctively damn sure that won't carry the current.
>
>Another option is some toroidal transformer, and ignore the existing
>winding and put more turns on it. Ebay is full of vintage transformers
>but most are pretty big.
>
>I was going to wind the TN16 with some turns and see what it does. I
>have an HP 3314 pulse generator and a power amp which can output 9V
>peak.
>
>The 8mH needs to be +/- 0.3mH. It was determined experimentally using
>this amazing thing
>https://www.ebay.co.uk/itm/224101042113
>but that obviously contains massive magnetics - it weighs about 1kg.
>
>I also have an LCR meter.
>
>I'd be grateful for any suggestions.

If you can scope a current waveform, you might get an idea of
what the most troublesome harmonics are, then see if a simple
LC filter will help. You'll need the C for any L to work.

RL