I really do hope I've learned a lot. That krunning - duger chart kind of put things in perspective.
I may indeed not be using this transformer to charge batteries however somewhere along the way I'll be needing transformers for a wind turbine I am building. As I am expecting large voltages i'll need some way of lowering them.
But that is a topic I might start here once my 3d printer is up and running again.

I have gotten a 0-30v 0-10A cc/cv lab power supply in the mean time. So I was thinking to use the transformer to quick charge with high amps the batteries to let's say 3.4 volts. and top off the voltage to 3.65 in a controlled and safe manner with the cc/cv

Also I think this project is a big success if we focus on the transformer alone certainly I do not feel defeated.

 

I know you’ve spent a lot of time/money on this project, and have probably learnt a lot along the way – but I have a feeling that you should read my post (#39), and accept defeat.

Even with a 5V, 70A supply, there is still work to do to control the battery charge current; if you can get it to work with a 5Vdc supply, you could revisit your home-spun transformer design at some later date.

You inspire me to reiterate that I also do not have high hopes for the transformer. I remain engaged in this thread out of curiosity. I have never seen or heard of anyone making a transformer out of loose iron filings and I think it's a cool, novel idea. That should not be taken as an indication that I think it will yield a good end result. I hope I have not given false hope by showing interest. I continue to be of the opinion that the relatively poor efficiency (relative to commercially available transformers or rewound MOTs) is better than anyone had a right to expect, and if it manages to charge a battery at high amps without burning down the house or poisoning the interior with deadly gas, then it can be called a "success" even if it wastes 75% of the energy purchased by TS. The rectifier design will be relevant whether or not TS uses the powder transformer or another transformer.

 

Very interesting - if slightly dangerous.

To answer this original question, what you need to do is add some "leakage inductance". This can be done by adding an air gap in the core but it's usually done by adding a magnetic shunt (with an air gap) across the magnetic circuit.
This is often done in arc welders, microwave ovens and neon sign transformers to limit the secondary current.
For a diagram and a picture of a transformer with an adjustable magnetic shunt see the bottom of this Wikipedia article: https://en.wikipedia.org/wiki/Leakage_inductance
Also see the attached picture.

It's only now I think I understood understand your suggestion. So it is not a variable air gap in the core you were talking ab9out but rather a variable air gap between the coils is it?

 

Air gap and leakage inductance...
https://en.m.wikipedia.org/wiki/Leakage_inductance

 

It's only now I think I understood understand your suggestion. So it is not a variable air gap in the core you were talking ab9out but rather a variable air gap between the coils is it?

No I think you had the right meaning of his comment the first time. Transformers in simple buzzbox welders have a screw adjustment that opens an air gap between the steel laminations (not the coils), thereby limiting current. But I think this is not what you need. This would be worth looking into if you were concerned about the transformer charging the batteries with too much current, but it seems you will have the opposite problem.

My apologies for butting in; I know the question wasn't addressed to me, but I couldn't help myself. I fought with myself about addressing the quoted post the first time around and when it looked the conversation moved on I let it go. But since you brought it back up... I hate to be directly contradictory, but it's my honest opinion that pursuing any air gap adjustment or magnetic shunt feature will be a waste of your time. Maybe worth revisiting if you move on to a transformer with steel laminations, but not for this iron powder experiment.

 

I'll be focusing the coming days on implementing safety features by adding a 4 amp circuit breaker before the variac and a 10 amp thermal fuse after the variac.

Also I am considering making it easier to try different coil configurations by making the core modular.
Currently making changes to the coils takes hours and makes it more likely one stops trying getting better results as soon as something useful (however inefficient) comes out of it.

Also

My apologies for butting in; I know the question wasn't addressed to me, but I couldn't help myself.

Please never stop. I could not have gotten this far without the butting in of people

 

I've gotten a 2A and 4A circuit breaker in the meantime and already installed the 2a one in between the power source and the variac in.

Now I am having more issues with inrush current than before. earlier every so often when switching on the system my 16A circuit breaker in the breaker cabinet would trip and I thought i'll deal with that later.

Can anyone recommend a suitable NTC thermal resistor? So many sizes I am not sure how to select a properly sized one.
If all else fails I'll replace the 2A circuit breaker with the 4A one but I'd rather keep the system on a tight leach.

Soon the 10A thermal fuse will arrive for installation in between the variac and primary coil.
Also soon i'll have the 3.6V 1F capacitor and then it's show time!! As in time to charge a battery.

The rectifiers I have not bought yet as first I want to give the one I have a fair chance to prove it self.

 

Can anyone recommend a suitable NTC thermal resistor? So many sizes I am not sure how to select a properly sized one.

Not me, sorry. I haven't used one in so long, I would have to re-teach myself about them and I don't have time. Hopefully someone else knows.

The rectifiers I have not bought yet as first I want to give the one I have a fair chance to prove it self.

What is the p/n of the one you have? You probably posted that information already but we are approaching 200 posts now and I wouldn't know where to go back and look for it.

 

Not me, sorry. I haven't used one in so long, I would have to re-teach myself about them and I don't have time. Hopefully someone else knows.

What is the p/n of the one you have? You probably posted that information already but we are approaching 200 posts now and I wouldn't know where to go back and look for it.

Nah, don't be sorry. I certainly do not expect and/or ask for people to dive into the theory books for me

But I always ask first in case someone knows it by heart which gives insight to everyone and saves a lot of time.

Anyway I could not find any details regarding the current full bridge rectifier other than what is posted on amazon. Those details are always to be taken with a grain of salt.
----
Technical data: output current: 150A; Repetitive Peak Reverse Voltage (VRRM): 1600V; Repetitive Peak Inverse Current (IRRM): 10mA; Insulation voltage (Viso): 2500VAC; VTM: 1.5V; Connections: 4; max. Transition temperature: 150 C
----


I could not wait though for all components to arrive and when ahead and tried charging a battery anyway in a controlled manner.

The good news is that the charging seem to work like a charm. The bad news is that I am far from done. So many things to do still to improve the very disappointing efficiency results.

I placed one battery outside so that it could do no damage if things went south.

Not sure if you can tell on this 800x600 image but the DC voltage reads 3.23VDC when the system is turned off.

By the way I also made sure there is no short circuit in the primary coil by measuring the resistance which was around 6 ohms. That means that the primary coil is around 280 meters in length like I reported. So no short circuit yet! fingers crossed it stays that way so but I am always very delicate when handling the transformer.
I kinda hoped there was a short as to have an easy fix for better results ;(

Now comes the sad part ;(

The efficiency in it's current state is around 11% before rectification and 9% after rectification if my calculations are correct.
(again sorry for full scale image posting but it's important that all details are clearly visible for people that want to contribute)

formula applied:
(output power / input power) * 100
((4,48×28,9)÷(134,3×8,71))×100 = 11,068319551 % before rectification
((3,65×28,9)÷(134,3×8,71))×100 = 9,017715706 % after rectification

After around 10 minutes of charging at 28.9 amps I turned off the system and found the battery now at 3.27VDC.
There were no temperature issues that I was aware of during the run.
Mind you that the capactiy of these cells is 280Ah so it takes a while for them to charge.
That means charging is working and I seem to have a valid proof of concept. Aiii!!!!!!!!!!
Also lifepo4 has a very particular voltage / SOC (state of charge) chart so I think these numbers are totally in order.

I will try a few things though to improve the amps before I start small scale coil/core configuration variation testing.

First i am going to try a better wire between rectifier out and battery. These car battery clamps were only for a quick test and I am hoping that the poor clamps are the result of the low (28.9) amp reading.

Second I will improve on the transformer container as to fully engulf the transformer in powder as I can see a little magnetic fields in the powder at the top which means I have to better enclose it with powder.

What I'd like to know is if the current full bridge rectifier can be improved upon by using the components suggested earlier.
If I read the char correct of those components a voltage drop of a bit more than 0.8 at 30 amps is to be expected.
With the current rectifier I see a voltage drop of 1.19 at around 30 amps if I am deducing it correctly.
So what to do here?

 

Your Amazon rectifier is a better unit than I had assumed. I don't think you will gain much by purchasing the components I suggested. I would just stick with what you have. Your
efficiency math looks right to me. But isn't this lower than before? I thought you were getting efficiency around 35% before. What changed?

 

Your Amazon rectifier is a better unit than I had assumed. I don't think you will gain much by purchasing the components I suggested. I would just stick with what you have. Your
efficiency math looks right to me. But isn't this lower than before? I thought you were getting efficiency around 35% before. What changed?

2 things changed.
First I placed the secondary coil inside of the primary. But that seems to have improved the efficiency to 39% when doing the tests we did earlier, ( the test we ran to deduce efficiency while the system still had the secondary coil on top of the primary)

As soon as we put this rectifier in the system things go south in terms of efficiency. At least that is how far I have gotten so far but as always I might be terribly mistaken.

 

ooops I made a huge brainfart.

This is only a proof of concept of a DIY transformer. That part is a success yes!

But this is still by no means suitable for quick charging batteries. I mean 30 amps is not at all 140.

The reason I am aiming high is that when I buy cells I need to be able to do a capacity test really quick before I loose my window of opportunity to complain.
Also when using a transformer in power generation projects it would be quite sad to throw away most of the generated power.

 

2 things changed.
First I placed the secondary coil inside of the primary. But that seems to have improved the efficiency to 39% when doing the tests we did earlier, ( the test we ran to deduce efficiency while the system still had the secondary coil on top of the primary)

As soon as we put this rectifier in the system things go south in terms of efficiency. At least that is how far I have gotten so far but as always I might be terribly mistaken.

It could be that your clamp meters are not measuring current accurately now that current is not sinusoidal. Are they RMS current meters? I suspect not since they don't say so; usually that is worth advertising. But that is quite a large disparity to be solely attributed to non-RMS measurements. I think. Maybe I'm wrong. Maybe this is totally predictable. I'm not sure, as I've been using RMS current clamp clamps for forever now and I don't remember how the others behave.

 

powder details in case not easily found on their website
-----
It is here the world famous CUT 150 powder
The purity is at least 99% (proven by analysis)
Sieve size distribution: <63µm 48,4% / >63µm 31% / >100µm 20,4% / >160µm 0,2%

Chemical analysis, weight %
Iron, >99.0%
Carbon, % 0.05 Leco C/S
O (H2-loss), % 0.60 DIN ISO 4491-2
Flow time s / 50g: 32.5
Bulk density: 2.79 g/cm³
Melting point: 1535 degrees
Boiling point: 2750 degrees
Chemically Stable
The particle shape is spherical
Manufacturing process: Air atomized

Generally, Fe 6% Si is used for transformers.
Bulk density =2.79g/cm³ is interesting.
Transformers generally achieve a stacking factor of >0.9, which would imply a density of >7g/cm³ (iron is 7800kg/m³)
if your bulk density is that low, then there is a lot a air gap, which would mean a high reluctance.
Hopkinson’s law would therefore give a very high magnetomotive force for a given level of flux, which would imply a low magnetising inductance, a high primary current and a very inefficient transformer.
With a spherical particle, the best packing density you could hope for is π/(3√2) = 0.74, which would be a poor transformer.
True, there would be very low eddy current losses, but they would pale into insignificance compared to the losses caused by the high primary current.

 

Generally, Fe 6% Si is used for transformers.
Bulk density =2.79g/cm³ is interesting.
Transformers generally achieve a stacking factor of >0.9, which would imply a density of >7g/cm³ (iron is 7800kg/m³)
if your bulk density is that low, then there is a lot a air gap, which would mean a high reluctance.
Hopkinson’s law would therefore give a very high magnetomotive force for a given level of flux, which would imply a low magnetising inductance, a high primary current and a very inefficient transformer.
With a spherical particle, the best packing density you could hope for is π/(3√2) = 0.74, which would be a poor transformer.
True, there would be very low eddy current losses, but they would pale into insignificance compared to the losses caused by the high primary current.

Thanks for the headsup. I will stop hoping getting better charging results than 39% using the current coil/core configuration.
In the meantime I have determined that the rectifier is crap. the terminals are wobbly and are about to let go.
Changing the car charge cable and clamps with 35mm2 wire and proper clams did change the efficiency to 13 % though after rectification and am now able to charge the battery at 47.7 amps.
I am going to have another hard look at rectifiers and come up with a suggestion to improve on that component of the system

 

I am going to have another hard look at rectifiers and come up with a suggestion to improve on that component of the system

Too bad I could not find anything better than the earlier suggestion so I ordered 2 of those and now we wait ;(
But along with that order there will be a couple of inrush current limiters so let's see how far I get with those as well.

 

Given the rather unfavorable composition of the powder I am looking again for laminated soft steel sheets.
The fact is that I tried earlier to find laminated soft steel sheets but I could not find large ones. So I ended up trying with powder never realizing what a poor choice I made.
So if anyone knows where i can purchase large sheets, or thick strips or anything for that matter I will also get that and then build another transformer. See how that goes.
Especially for power generation application it is important not to waste too much energy.

That is not to say I am done with the powder just yet. I fully intend on going small scale and try endless permutations of coil and core and see what the practical maximum in terms of efficiency is.
I mean powder is soo easy to work with. and reusable to make different shapes over and over.

 

https://ascocomponents.co.uk/laminations/

 

I’m impressed that you have managed to get your home-spun transformer to charge the battery, albeit at a very poor efficiency.

I suspect that the main reason for the poor efficiency is that the iron filings do not make a good a magnetic circuit, compared with solid steel/iron – and therefore does not induce as high a secondary current.

Before you give up on the iron filings, I’d try increasing the number of secondary turns, this in turn should need a lower primary voltage to achieve the required secondary voltage – which may increase the efficiency.

 

Given the rather unfavorable composition of the powder I am looking again for laminated soft steel sheets.
The fact is that I tried earlier to find laminated soft steel sheets but I could not find large ones. So I ended up trying with powder never realizing what a poor choice I made.
So if anyone knows where i can purchase large sheets, or thick strips or anything for that matter I will also get that and then build another transformer.

MICROWAVE. OVEN. TRANSFORMERS.
JEEZUZ.
If I ever meet you I will shake your hand and then immediately hit you over the head with a MOT.