DIY current transformer for measuring primary current?

[Daniel Uhrenholt]

Hi All,

I don’t know if anyone remember me from a decade ago

The thing is that a decade ago I stopped making Tesla Coils, and sold my differential probe and Pearson current transformer to Mads Barnkob. My problem right now is that I wanted to show my 5 year old son my bipolar DRSSTC, that have a permanent space in the living room, make some sparks and arcs for the first time in 5 years. Bang it said, and the halfbridge exploded.

Halfbridge replaced, but can only measure the gates at the moment.

I have a differential probe on the way in the mail, but I need a current measuring transformer that works in the 100-200 kHz range just to see what is going on. I miss my old Pearson, but can’t afford a new one at the moment, bought a house and have two kids now.

Any suggestions on making one with an okay bandwidth?

Cheers,

Daniel Uhrenholt

Edit: can anyone move this thread to the DRSSTC area?

[Hydron]

Looks like someone has moved it before I got there.

As for a DIY CT, to do what you want to is probably one of the easiest DIY probes to build. You won't get as good e-field shielding as the Pearson (so it will pick up noise from fast switching transitions) or good low-frequency response, but in the area of interest (100-200kHz) a cascaded DIY CT will work extremely well.

I've posted a thread previously on this (https://highvoltageforum.net/index.php?topic=914.0) but the short story is that you need a couple of ferrite toroids (specification un-critical as long as they're not too low-Al) some wire and a resistor or two The first one in my thread was literally made of junk-box parts (I'm using 4 ferrite cores for a single input followed and 3 independent secondary outputs - for monitoring plus UD2.x driver inputs).

[Daniel Uhrenholt]

Thanks for the link

I will take a deeper look at it later today😊

I know how to make a CT, but making one with a fine sine output is another thing... I thought that someone had figured something out by now

I remember Finn Hammer and I discussed a design of a current measuring transformer back in the days, and a PCB was thought about. Sitting with my iPad is making it a bit hard to do a drawing, so let’s see if anyone get my point. The idea was to make a pcb to mount a toroid on, and have each winding mounted on the pcb and have a resistor at each winding on the back, so the impedance would be dived over the windings. A 15 year old discussion

I have to look at some ferrite toroid with a ID of 50mm, so I can get cables and stuff through it

[Hydron]

For just drsstc work I think a complicated design is unnecessary - the relatively small number of windings on each core needed to achieve a practical ratio in a cascaded CT means that parasitics aren't such a worry at the frequencies of interest.

If you need a wide frequency range then it's another story - my Pearson 2788s are specified for use over 6 orders of magnitude (300Hz to 300MHz)! In this case trying to do a DIY PCB based CT was something i also thought of trying, but in the end I got lucky on eBay a few times and collected a number of Pearson CTs (5 in total, averaging about 60£ each)

Having said all that, I'd definitely be interested in the results of a PCB based approach if someone tries it - is very inexpensive these days to give it a go.

[Daniel Uhrenholt]

[ quote author=Hydron link=topic=1277.msg9406#msg9406 date=1603554169]
If you need a wide frequency range then it's another story - my Pearson 2788s are specified for use over 6 orders of magnitude (300Hz to 300MHz)! In this case trying to do a DIY PCB based CT was something i also thought of trying, but in the end I got lucky on eBay a few times and collected a number of Pearson CTs (5 in total, averaging about 60£ each)
[/quote]

I don’t remember the model or ratio on the Pearson CT, maybe Mads can look at it and post it here😊 I would like to make something like that... I think it was around 0.1V / A

Everything over 20MHz is overkill for what I’m working with these days

The price for a Pearson CT on eBay right now with shipping, import tax and handling to Denmark from the US is £300+. That is the price for daycare for one child 1 month😂 My wife would kill me if I do that

I might try the PCB way, with this Corona winther with no social life at all, and lockdowns around the world. But I will need some guidance in finding the right ferrite core for the job. I buy special components from RS components, If anyone would help an electronic demented guy from Denmark to find a good core. If I decide to do the PCB, I will make it accessible online.

I need to get my head spinning again with this, the last many ears I’ve been coding strategies/macros for CNC machines, and haven’t touched any analogue electronics at all.

Cheers, Daniel

[Hydron]

I would maybe look at some of the nanocrystalline cores, e.g. https://www.vacuumschmelze.com/Nanocrystalline-Material
Also if you mount it on a PCB for placing distributed shunt resistors, using solderable enamel wire may make it much easier as you could use a single length and tack it onto a pad underneath the wire rather than terminating every time you want to place a resistor.

As for Pearson prices, you basically need to be patient, they aren't common in the UK but occasionally came up over a period of 3-4 years:
- The first one of mine was in bad shape but went for <50GBP in an ebay auction (this is the one that needed BNC surgery)
- A couple I found via a saved search on different days from the same seller (sadly the only ones he had) - Mads saw them first but I grabbed them before he got a shipping quote (or knew he was interested), both around the 100GBP mark
- The last two (tiny model 2877s) came from a US seller who had a big pile of them marked as model 6298 (Pearson confirmed they were 2877s with extra testing), I made an offer for approx. 100USD for two and got them sent to a US based friend and picked them up nearly a year later (and might have forgotten to declare them when flying back...).

Mads also picked up a massive clamp-on one a year or two back for under 150 EUR, he may have an amusing picture of it in use.

[Mads Barnkob]

I feel guilty for you not being able to pick up where you left

Maybe this good old DIY CT resource is enough for you? https://interferencetechnology.com/the-hf-current-probe-theory-and-application/

I still use the differential probe, so I am not letting go of that.

But was it a Pearson model 110 that I bought from you? As Hydron mentioned I did fetch a clamp on with same 0.1V ratio as the 110, so we could maybe work about a buy-back

Picture related to both Pearson clamp-on and introducing wife to a new kind of jewelry

 

[Daniel Uhrenholt]

But was it a Pearson model 110 that I bought from you? As Hydron mentioned I did fetch a clamp on with same 0.1V ratio as the 110, so we could maybe work about a buy-back

Lets talk abort that over PM Odd thing buying stuff back after 8 years or so

I feel guilty for you not being able to pick up where you left

Come on, I have been looking at your website from time to time over the years, and you do stuff that will take me years to figure out myself you have a background in electronics and I in mechanics/tooling For me making sparks with Tesla Coils wasn’t the funniest part of coiling, it was the whole creative process and building stuff that no one had done before in this hobby. And challenging my craftsmanship as a toolmaker, learning new things all the time.

Nice necklace, is there a model that fits on fingers? The woman of the house keeps talking about a wedding ring

Hydron:

If you ascendency break a Pearson CT and disassembly it, that foil material is the core off it or something similar I guess.

I might take the challenge of making the PCB CT just for fun, and use a Pearson CT for calibration I have always dreamed about having an permanent CT for measurements in some of my old creations, but Pearson was not an option, and will not be before the kids move out

Cheers

Daniel

[Hydron]

If you ascendency break a Pearson CT and disassembly it, that foil material is the core off it or something similar I guess.

Yeah this is why I suggested it. I have got a similar core salvaged from a massive 3-phase CM choke Wuerth gave me as a sample. Some people on EEVBlog forums have also had a play around using them for injection transformers (high Al = fewer turns needed = better parasitics).

[plasma]

The terribly twos, remembered lifting up the head of the cot to stop gas.
Maybe you should wait another 5 years, just saying.

[Daniel Uhrenholt]

The terribly twos, remembered lifting up the head of the cot to stop gas.
Maybe you should wait another 5 years, just saying.

Edit: misunderstood the comment :-) I plan on teaching the oldest programming around 3 year%u2019s from now.

WTF?

Maybe the name of the thread should have been DIY Current Measuring Transformer, and I a little more specific about what I want ask about. Being a educated toolmaker, my electronics and programming skills are self learned, and not part of my education.

I may be a bit rusty with this at the moment... But My work and projects with Finn Hammer years ago, lead to the design of the first phase lead DRSSTC drivers, that everyone use in their DRSSTCs today... Finn and I are both Educated Toolmakers, and do not have any degree in electronics. But still managed to make things at a rather high level in Tesla Coiling, High Voltage and power electronics desig. Because of our craftsmanship, mechanical knowhow, out of the box thinking, and a drop of spark madness.

Have you contributed with anything to the Tesla Coiling Community that Im not aware off?

First phase lead DRSSTC in the world, where I was a part of building.

https://4hv.org/e107_plugins/forum/forum_viewtopic.php?p=1&id=68820

My coil with Remote tuning of primary coil on the fly when coil is running. The second phase lead driven coil ever...

https://4hv.org/e107_plugins/forum/forum_viewtopic.php?p=1&id=74660


Well back to CT:

I have tried the DRSSTC CT as suggested many years ago, two cores and a roughly 1000:1 output. Worked ok and I still have it somewhere, but the waveform looked awful in the start of a burst, compared to a Pearson monitor, just as Hydron mention in his thread. Core material was N87 - blue coloured toroid from Epcos

Being low on equipment at the moment, and need to fine tune a small Bipolar DRSSTC with a resonant frequency in the 170-180 kHz range, asking here seemed the way to go. Maybe someone had made a design for a measuring CT with good frequency response over the years, that I could use instead of paying a lot of money getting an Pearson from the US, or use the time on trying to design one myself. Or just links to websites containing information on designs and calculations.

A 1000:1 CT is in the high end for this coil because of the relative low current flowing in the primary, the output voltage of the CT is on the low side to really see whats happening at the first cycles because of the noise and voltage level. A 100:1 would be a better choice, and use one core with a large diameter to get some distance between the windings to reduce inter-winding capacitance to get a better result, and make some kind off shielding on it too. I still like the idea dividing the burden resistor with the windings of the CT and have a resistor at each winding I dont know if it will gain much on performance, but the overall neatness of the CT will be eye catching.

Cheers, Daniel

Edit: changed to Thumper thread at 4hv, it linked to my coil

[davekni]

"My coil with Remote tuning of primary coil on the fly when coil is running. The second phase lead driven coil ever..."

Looked back through your 4HV link.  A very impressive coil!  I especially like your connection from IGBT bridge to bulk caps.  Nice parallel planes with very low inductance.  (My skills are opposite yours.  My background is physics and electronics, with very little mechanical ability.  My DRSSTC base is plastic milk crates.)

Do I understand correctly that your on-the-fly tuning is adjusting coupling factor?  I'm working on a project to dynamically tune the primary resonant frequency, increasing primary MMC capacitance during individual enable pulses, so primary frequency will track secondary.

Concerning current transformers, the one I built for my DRSSTC would probably meet your needs.  I used an E80 ferrite core rather than a toriod.  Started with a single layer winding of 40 turns of 24AWG magnet wire.  Added just over one turn of copper foil, insulated at the overlap (no shorted turn), and connected to ground as a shield.  Then added one turn of copper 0.2mm foil for the primary (actually two layers of 0.1mm foil).  Tape between each layer for insulation.  Both the primary and secondary windings are almost the full width of the E80 core pair center leg, so leakage inductance is low.  The secondary of this transformer feeds two conventional toriod transformers, each with 2-turn primary and 25-turn secondary, then to 5-ohm burden resistance (1V/100A).  I haven't explicitly measured frequency response, but scope plots from operating my DRSSTC look quite clean.  Works well to 2600A.  Designed for 3500A eventually.  (I also added windings on the two E-core outer return legs, but these don't appear to be necessary.  This was just in case there was enough leakage inductance to saturate part of the core at 3500A.)

Concerning plasma's comment, I'm guessing that he was referring to the ages of your children, that parenting young children takes too much time to allow for coiling.  Your accomplishments are amazing.