Does anyone know how one can independently measure the thickness of enamel and/or enamel permittivity of enameled wire?
Of course, if one knows either one of these two qualities, the other can be determined by the salt water immersion test (measure capacitance between the wire and the water.) A large scale transformer manufacturer who buys wire in hundreds of kilometres, the manufacturers would no doubt provide such data. But for me, a one man semi-hobbyist/semi consultant, that won't happen. The IEC published a standard, 60317, but manufacturers of wire readily available seem to ignore it.
One cannot just use a micrometer on the OD because the tolerance on copper diameter is large compared to enamel thickness. I tried various means of removing enamel, such as scrapping, heating with a gas flame then scrapping, but got very inconsistent results. Common solvents such as gasoline, paint stripper, etc have no effect at all on the enamel.
Note: I posted this question a few weeks ago, but the thread was hijacked by people, obviously not involved in inductor/transformer design or making such, seeking to convince me I don't need to know. If I didn't need to know, I would not have asked the question.
In a machine-wound closed packed winding, capacitance is determined by both the enamel and the surrounding air or impregnant. One needs to know both thickness and pemittivity of the enamel because the electric flux is concentrated where the tuns touch - the degree of concentration is dependent on the ratio of these quantities.
some thought there is no mathematical formula one can use to predict self capacitance. But there is - for example Massarini's method, or Koch's method.
Of course, if one knows either one of these two qualities, the other can be determined by the salt water immersion test (measure capacitance between the wire and the water.) A large scale transformer manufacturer who buys wire in hundreds of kilometres, the manufacturers would no doubt provide such data. But for me, a one man semi-hobbyist/semi consultant, that won't happen. The IEC published a standard, 60317, but manufacturers of wire readily available seem to ignore it.
One cannot just use a micrometer on the OD because the tolerance on copper diameter is large compared to enamel thickness. I tried various means of removing enamel, such as scrapping, heating with a gas flame then scrapping, but got very inconsistent results. Common solvents such as gasoline, paint stripper, etc have no effect at all on the enamel.
Note: I posted this question a few weeks ago, but the thread was hijacked by people, obviously not involved in inductor/transformer design or making such, seeking to convince me I don't need to know. If I didn't need to know, I would not have asked the question.
In a machine-wound closed packed winding, capacitance is determined by both the enamel and the surrounding air or impregnant. One needs to know both thickness and pemittivity of the enamel because the electric flux is concentrated where the tuns touch - the degree of concentration is dependent on the ratio of these quantities.
some thought there is no mathematical formula one can use to predict self capacitance. But there is - for example Massarini's method, or Koch's method.
Perhaps gallium can be used instead of salt water?
Also if you have an accurate micrometer you could try measuring the wire thickness before and after burning off the insulation? Burn it hot enough and only the copper will be left I think - orange heat or so.
Also if you have an accurate micrometer you could try measuring the wire thickness before and after burning off the insulation? Burn it hot enough and only the copper will be left I think - orange heat or so.
Salt water works just fine. Capacitance bridges ignore the resistance in the water. About a litre is needed to comfortably immerse a metre of wire, and you need about that much to get sufficient capacitance for a decent reading (a few hundred pF). It costs about nothing and is completely non toxic. When you are done you can pour salt water down the kitchen sink. If gallium is used, about $400 worth is needed, plus shipping. It is somewhat toxic. And you can't poor it down the sink. You have to pay to have it disposed of.
Burning the enamel off is not simple. I tried doing that with a gas flame. If it is hot enough to get rid of the enamel, oxygen from the air oxidises the copper and changes its diameter - you can actually get a higher micrometer reading than over the untreated enamel. Charing the enamel black leaves the enamel still strongly clinging to the copper - it won't rub off.
Burning the enamel off is not simple. I tried doing that with a gas flame. If it is hot enough to get rid of the enamel, oxygen from the air oxidises the copper and changes its diameter - you can actually get a higher micrometer reading than over the untreated enamel. Charing the enamel black leaves the enamel still strongly clinging to the copper - it won't rub off.
Some discussion on enamel stripping at: https://www.eevblog.com/forum/beginners/what-dissolves-removes-magnetic-wire-enamel-coating/
Thanks. Doesn't help much though.
Almost all of the posts assume the reason for getting the enamel off is to solder the wire. Thus they recommend just holding a soldering iron to the wire along with some solder, or using a solder pot - which of course works -the enamel is designed that way. But it leaves the wire coasting with solder, which changes its diameter.
Others suggest mechanical abrasion (eg sand paper), which removes some of the copper.
One guy suggested methylene chloride (paint stripper) but that simply doesn't work. If given time and heat, it will attack copper.
A couple suggested a commercial wire stripping tool. These are grinders and will presumably remove some copper. (there are also wire strippers that use scraping blade method)
One guy suggested heated aspirin, which is available in pure form, but is very unlikely to work, as wire enamel is supposed to be resistant to weak acids.
One guy suggested conc sulphuric acid, which will attack the copper, converting it to copper sulphate.
Has anybody tried aspirin powder?
Almost all of the posts assume the reason for getting the enamel off is to solder the wire. Thus they recommend just holding a soldering iron to the wire along with some solder, or using a solder pot - which of course works -the enamel is designed that way. But it leaves the wire coasting with solder, which changes its diameter.
Others suggest mechanical abrasion (eg sand paper), which removes some of the copper.
One guy suggested methylene chloride (paint stripper) but that simply doesn't work. If given time and heat, it will attack copper.
A couple suggested a commercial wire stripping tool. These are grinders and will presumably remove some copper. (there are also wire strippers that use scraping blade method)
One guy suggested heated aspirin, which is available in pure form, but is very unlikely to work, as wire enamel is supposed to be resistant to weak acids.
One guy suggested conc sulphuric acid, which will attack the copper, converting it to copper sulphate.
Has anybody tried aspirin powder?
Perhaps one could estimate the average diameter of the conductor based on a given length using a milliohm meter with sufficient precision, assuming the wire is made of high purity copper hence having a known ρ.
Gee, I can't see easy success in measuring resistance to find diameter. Resistivity is dependent on work hardening, which copper is very susceptible to. Even just unwinding wire from the spool it is supplied on produces noticeable hardening in the heavier gauges. One would need to anneal it at 400 C before testing. Copper has the usual metal positive temperature coefficient of resistance (circa 0.4% per C), so one would need to have the wire in a precision oven for testing. It happens I do have such an oven.
Resistivity of commercial copper at standard temperature is only given to 4 places. That leads to uncertainty in diameter of the same order as enamel thickness.
One meter of 1 mm diameter wire has a nominal resistance of 3.07 milli-ohms -not easy to measure to the 5 places necessary without causing significant self heating.
Worth some further thought though.
Resistivity of commercial copper at standard temperature is only given to 4 places. That leads to uncertainty in diameter of the same order as enamel thickness.
One meter of 1 mm diameter wire has a nominal resistance of 3.07 milli-ohms -not easy to measure to the 5 places necessary without causing significant self heating.
Worth some further thought though.
Thanks. 2 posters there say it works. I will investigate further, and find out where I can locally buy aspirin powder.
Salicylic acid is a fairly strong acid, pH 3, it is used to dissolve warts amongst other things. Heating aspirin may breakdown salicylate to salicylic acid?
I just flamed some SWG32 wire to orange heat (nominal diameter 0.28mm), and it measures 280µm, whereas 305µm with insulation intact, implying 12.5µm insulation thickness. However even a good micrometer will have some uncertainty around 1µm, so its not a super-precise measurement (also wire-drawing dies may not be accurately circular so there can be variation due to wire orientation too).
I can see process variation in manufacture leading to insulation thickness varying along the same piece of wire anyway. And just knowing the thickness without the dielectric constant doesn't get you the capacitance anyway.
Mark, pH 3 is a weak acid. However, as of now three people have said it works, so I will see if I can get it when the shops open on Monday.
When you heated your wire to orange heat, what was the appearance when it cooled down - bright copper, discoloured, black, or what?
An enamel thickness or 12 um is suspect, since IEC-60317 mandates a minimum enamel thickness of 18 µm for 0.28 mm wire (grade 1 wire, other grades have thicker enamel). You must have been lucky to get a sample of wire measuring exactly 280 µm.
Re your last paragraph. as i said in my initial post, if you know either the thickness or the dielectric constant, the salt water immersion capacitance test will give you the other, and then you have the inputs required for Massarini's formula which will give you the capacitance between 2 adjacent wires in a hexagonally packed bundle.
I would be doing well to read my micrometer's vernier scale to better than 2 µm, but that would be good enough for my purposes. The interstice space reduces the influence of the enamel on capacitance to some extent.
You are correct about does not being precisely circular, but it should be good enough to take 2 readings at right angles and average them.. One cannot generally measure with the micrometer jaws directly on the wire, as tiny kinks in the wire, not visible to the eye, will make the reading high. This is solved simply by using the cross-wire technique - two accurately ground and checked steel wires between the micrometer jaws and the wire under test, and at right angles to the wire under test, so that the contact is at one point only, and not across the width of the flat faces of the jaws.
When you heated your wire to orange heat, what was the appearance when it cooled down - bright copper, discoloured, black, or what?
An enamel thickness or 12 um is suspect, since IEC-60317 mandates a minimum enamel thickness of 18 µm for 0.28 mm wire (grade 1 wire, other grades have thicker enamel). You must have been lucky to get a sample of wire measuring exactly 280 µm.
Re your last paragraph. as i said in my initial post, if you know either the thickness or the dielectric constant, the salt water immersion capacitance test will give you the other, and then you have the inputs required for Massarini's formula which will give you the capacitance between 2 adjacent wires in a hexagonally packed bundle.
I would be doing well to read my micrometer's vernier scale to better than 2 µm, but that would be good enough for my purposes. The interstice space reduces the influence of the enamel on capacitance to some extent.
You are correct about does not being precisely circular, but it should be good enough to take 2 readings at right angles and average them.. One cannot generally measure with the micrometer jaws directly on the wire, as tiny kinks in the wire, not visible to the eye, will make the reading high. This is solved simply by using the cross-wire technique - two accurately ground and checked steel wires between the micrometer jaws and the wire under test, and at right angles to the wire under test, so that the contact is at one point only, and not across the width of the flat faces of the jaws.
There are industrial solvents capable of stripping off almost any cross-linked polymer, even the epoxy of FR4. Panasolve 160 is an example that comes to mind, but it's not the only one.
They are completely inert for metals
They are completely inert for metals
Googling <panasolve 160> turned up nothing. There is a Panasolve that is a solvent for engine oils. What company supplies Panasolve 160?
Immersing a meter of wire in salt water to measure the capacitance sounds like a good idea. If you are paranoid about the cut end of the wire, dip it in polyurethane varnish (just a tich) to seal the cut end. If you are talking about standard magnet wire, well, that has a composite coating of polyurethane for solder strippability, with an outer coat of nylon for abrasion resistance. That particular coating option has several trade names, including Nyleze and Solderon. One of the most valuable lessons I leaned form the techs at my first job was that magnet wire was solder-strippable, so you could use a std soldering iron or a solder pot to strip the magnet wire coating and solder it to a lug or some such termination.
We routinely use a solder pot at my current job to terminate magnet wire to pins on transformer bobbins. Many of the bobbins we use are molded using phenolic (thermosetting) resin, so they can stand up to the heat from a solder pot. Standard bobbins molded from thermoplastics like Nylon and PBT (both thermoplastics capable of being used with STD injection molding machines) fare less well with the heat stress from the solder pot. Magnet wire exists with both single and double coating options. The double coated wire will stand up to the relatively rough handling it can get with an automated winding machine. Figure out which type/thickness of wire coating you are measuring before you do the deed.
We routinely use a solder pot at my current job to terminate magnet wire to pins on transformer bobbins. Many of the bobbins we use are molded using phenolic (thermosetting) resin, so they can stand up to the heat from a solder pot. Standard bobbins molded from thermoplastics like Nylon and PBT (both thermoplastics capable of being used with STD injection molding machines) fare less well with the heat stress from the solder pot. Magnet wire exists with both single and double coating options. The double coated wire will stand up to the relatively rough handling it can get with an automated winding machine. Figure out which type/thickness of wire coating you are measuring before you do the deed.
Last edited:
Panasolve seems to be gone (for environmental reasons maybe), but you can still find alternatives:
https://www.indiamart.com/proddetail/copper-wire-enamel-remover-chemical-22491048488.html?mTd=1
https://www.indiamart.com/proddetail/copper-wire-enamel-remover-chemical-22491048488.html?mTd=1
Best to find something with an MSDS... The definition of dodgy is a large bottle labelled 'paint stripper chemical'!!!
Paint stripper - whose main active ingredient is methylene chloride - doesn't work on the usual solder strippable magnet wire coating. A dip in a solder pot cleans off the insulation and tins the wire, too.
There are some horrible chemical concoctions that work for stripping heavy polythermaleze wire, used to wind motors, where the possibility of abrasion is high (and operating temperatures are high, too). Those stripping chemicals have to be cleaned off immediately afterwards - perhaps they start to corrode the underlying copper.
There are some horrible chemical concoctions that work for stripping heavy polythermaleze wire, used to wind motors, where the possibility of abrasion is high (and operating temperatures are high, too). Those stripping chemicals have to be cleaned off immediately afterwards - perhaps they start to corrode the underlying copper.
Let's face it: if something is capable of loosening the bonds of a cross-linked resin like polyimide, it is not going to do you any good. Even dichloromethane, a relatively "gentle" stripper is not going to do you any good, even at low doses.
You have to know what you want, and take suitable precautions if required (all the time in fact with this class of products)
You have to know what you want, and take suitable precautions if required (all the time in fact with this class of products)
Methylene chloride is no fun - it burns like the hinges of Hell when it contacts unprotected skin - ask me how I know....