### Author Topic: 20W 50Hz isolation transformer  (Read 687 times)

0 Members and 1 Guest are viewing this topic.

• Frequent Contributor
• Posts: 822
• Country:
##### 20W 50Hz isolation transformer
« on: April 25, 2019, 04:35:45 pm »
Hi,

I want to design and build an Isolated 380Vac to 220vac single phase(50Hz) 20W transformer, the main thing about it is that it should withstand 6Kv isolation for burst tests, I want to short the input and outputs and test the 6Kv isolation. The windings also should withstand 6Kv isolation to the core.

The problem with this transformer is that I have lack of information, first of all I know these formulas from the high school.

V1/V2 = n1/n2

But How should I calculate the core size, which is an Iron EI core, what’s the wire sizes? What’s the primary winding turns?

I have another 220 Vac to 220 Vac 200W transformer Which I have bought many years ago. I measured it’s primary winding @ 100Hz and it’s in the order of 275mH.

What should be the size of insulation and bobbin thickness to pass the 6kV isolation? Do we have a tool for calculations? Any Ideas are very welcome.

Thanks
You can order parts from www.ASiDesigner.com
we are a wire-based company

#### soldar

• Super Contributor
• Posts: 2002
• Country:
##### Re: 20W 50Hz isolation transformer
« Reply #1 on: April 25, 2019, 06:49:02 pm »
Some very, very, and I mean *very* rough rule of thumb calculations:

I assume 20 VA rather than 20W. This is a very small transformer. The core section depends of the characteristics of the core but I would start our with, say, 5 cm2.

At 3A / mm2 the 380V primary can use 0.15 mm diameter wire and the 220 V secondary 0.2 mm.

Say 3800 turns on the primary and 2300 on the secondary. That is my first rough iteration. Calculate the bobbin section and if not enough go up one size.

If you want more precise calculations we would need to know magnetic characteristics of the core, etc.

Regarding 6KV isolation, I suggest this is for someone who has more knowledge and experience. If I had to build it myself I would go for separate bobbins on separate branches of a C core as that would make things much simpler. If you wind both on the same bobbin you really need to be careful and know what you are doing.

In air you would want to keep at least 6 mm separation and I would go for more if possible. With other materials it can be less but you would need to make absolutely certain there are no gaps or cracks. As I said, separate bobbins on a C core make it much easier. Still you need to study the bobbin material and insulating characteristics.

« Last Edit: April 25, 2019, 07:12:40 pm by soldar »
All my posts are made with 100% recycled electrons and bare traces of grey matter.

• Frequent Contributor
• Posts: 822
• Country:
##### Re: 20W 50Hz isolation transformer
« Reply #2 on: April 26, 2019, 05:45:19 am »
Thanks for the feedback

Quote
The core section depends of the characteristics of the core but I would start our with, say, 5 cm2.
So what is the length and width and height of the core? is there any standard core available?

How did you calculate the turns, the core is Iron,does a 4mm ABS  bobbin withstand 6kV? how should I calculate it?

Also is there a certain way to turn the wires, I mean for the isolation?
You can order parts from www.ASiDesigner.com
we are a wire-based company

#### soldar

• Super Contributor
• Posts: 2002
• Country:
##### Re: 20W 50Hz isolation transformer
« Reply #3 on: April 26, 2019, 07:35:10 am »
Why do you want to design and build this yourself? Is it only to save some money?  Because, to be honest, I get a feeling you would be much better off just ordering it from a professional manufacturer. They can design and build in no time. They have all the supplies they need and plenty of expertise.

You will have much difficulty in buying only what you need and will have to buy much which will be wasted. But, most important of all, while the design is not hugely complicated for someone experienced, it is really outside of your level right now. If you really want to do this I suggest you get started learning how to design and build transformers. It is not extremely complicated but there are many small things. What laminations do you have access to? What are their magnetic and mechanical characteristics? Where are you going to buy all your supplies? If it were for me I would overbuild it just to be extra safe.  You want extra voltage isolation and you need to get the characteristics of the materials you will use. Where would you get your supplies?

What is the purpose of this transformer? Should it have some protection?

Frankly, I do not feel comfortable designing a 380 / 230 V transformer with 6KV isolation for myself, and much less for someone else who is a beginner. If you just want the transformer I suggest you buy it from a professional. If you want to learn I suggest you start by reading -- a lot. And then start with something more modest.

Note that professionally built transformers are going to be much better built. For instance, they will vacuum and fill with varnish which seals and holds everything in place. You are not going to be able to do this as well. I mean you can just soak it in varnish but the result will not be the same. They have winding machines which you don't. It is easy to damage the insulation on the wire if you are not extremely careful. You could end up wasting a lot of time and supplies ... and maybe causing some damage.
All my posts are made with 100% recycled electrons and bare traces of grey matter.

#### Circlotron

• Super Contributor
• Posts: 1535
• Country:
##### Re: 20W 50Hz isolation transformer
« Reply #4 on: April 26, 2019, 07:46:14 am »
A 380V 3-phase supply measures 220V from any phase to neutral. If you have the neutral wire available, do you really need a transformer? What is the function you want to achieve?

#### soldar

• Super Contributor
• Posts: 2002
• Country:
##### Re: 20W 50Hz isolation transformer
« Reply #5 on: April 26, 2019, 08:06:26 am »
You must have missed the 6 KV isolation part.
All my posts are made with 100% recycled electrons and bare traces of grey matter.

#### Zero999

• Super Contributor
• Posts: 12730
• Country:
• Hero999
##### Re: 20W 50Hz isolation transformer
« Reply #6 on: April 26, 2019, 06:02:56 pm »
If you measured 275mH, then the transformer is either bad or it's a measurement error. Perhaps the frequency was 100kHz, rather than 100Hz?

Why not just buy one? 6kV of isolation is on the high side, but should be doable. Is it 6kV peak or RMS?

6kV peak between windings is no problem and the isolation to the core requirement can be avoided by isolating the core from earth.

For example, this standard transformer has 5kV RMS (7kV peak) rated insulation between the primary and secondary.
https://uk.rs-online.com/web/p/din-rail-panel-mount-transformers/1235698/
https://docs-emea.rs-online.com/webdocs/1603/0900766b81603bbf.pdf

#### soldar

• Super Contributor
• Posts: 2002
• Country:
##### Re: 20W 50Hz isolation transformer
« Reply #7 on: April 26, 2019, 06:31:25 pm »
Here's a good intro to get you started.
https://ludens.cl/Electron/trafos/trafos.html
All my posts are made with 100% recycled electrons and bare traces of grey matter.

#### soldar

• Super Contributor
• Posts: 2002
• Country:
##### Re: 20W 50Hz isolation transformer
« Reply #8 on: April 26, 2019, 09:57:48 pm »
6kV peak between windings is no problem and the isolation to the core requirement can be avoided by isolating the core from earth.

I would not isolate the core from earth, I would rather build up the bobbin walls if necessary.  By using a split bobbin of necessary thickness or building up the walls if necessary this should be no problem.

My uneasiness here is to have someone who is out of his depth building something that can be very dangerous.  For such small power and standard voltages I would just go out and buy one.  Building makes no sense.
All my posts are made with 100% recycled electrons and bare traces of grey matter.

• Frequent Contributor
• Posts: 822
• Country:
##### Re: 20W 50Hz isolation transformer
« Reply #9 on: April 27, 2019, 12:09:02 am »
Thanks soldar for sharing

Actually I should design it and send the instruction for manufacturing, it's my customer needs. also I have a 3D printer, so I can make a custom designed bobbin for the sample.

I have a variac too, so I do not think building the sample would be a huge step for me, I can have control on things.

Quote
If you measured 275mH, then the transformer is either bad or it's a measurement error. Perhaps the frequency was 100kHz, rather than 100Hz?

Why not just buy one? 6kV of isolation is on the high side, but should be doable. Is it 6kV peak or RMS?

It's  not a measurement error, I have measured it with my good Latron RLC meter,
Also the isolation is for a 6KV peak voltage, I have built a 3K boost converter before, with the diode,cap multiplier combination, I can increase it's output or build two of them and use it for testing the isolation.

You can order parts from www.ASiDesigner.com
we are a wire-based company

#### Benta

• Super Contributor
• Posts: 2051
• Country:
##### Re: 20W 50Hz isolation transformer
« Reply #10 on: April 27, 2019, 12:15:35 am »
I my opinion, you're wasting your time. Send the specs to a transformer manufacturer and he'll roll one for you just like that. He won't even calculate it, but just select from his own standard table of cores and winding specs.
And go for 400/230 V instead, then it's universal for many countries and will still work as 380/220.

#### soldar

• Super Contributor
• Posts: 2002
• Country:
##### Re: 20W 50Hz isolation transformer
« Reply #11 on: April 27, 2019, 12:57:47 am »
It's  not a measurement error, I have measured it with my good Latron RLC meter,

275 mH is too low and the no-load current would be too high, probably saturating the core. You need to review your measurements.

What happens when you measure straight resistance (Ohms)?

What happens when you connect it to 230 Volts? What current does it draw?

IMHO, if you want to learn then learn first and if you want a transformer then buy it but you are not equipped right now to complete this job safely and successfully.
« Last Edit: April 27, 2019, 03:06:06 am by soldar »
All my posts are made with 100% recycled electrons and bare traces of grey matter.

#### soldar

• Super Contributor
• Posts: 2002
• Country:
##### Re: 20W 50Hz isolation transformer
« Reply #12 on: April 27, 2019, 08:11:41 am »
Another interesting page: Calculating mains frequency power transformers

http://www.giangrandi.ch/electronics/trafo/trafo.shtml
All my posts are made with 100% recycled electrons and bare traces of grey matter.

• Frequent Contributor
• Posts: 822
• Country:
##### Re: 20W 50Hz isolation transformer
« Reply #13 on: April 27, 2019, 02:44:03 pm »
Quote
Another interesting page: Calculating mains frequency power transformers

http://www.giangrandi.ch/electronics/trafo/trafo.shtml

You can order parts from www.ASiDesigner.com
we are a wire-based company

#### Zero999

• Super Contributor
• Posts: 12730
• Country:
• Hero999
##### Re: 20W 50Hz isolation transformer
« Reply #14 on: April 27, 2019, 05:41:48 pm »
It's  not a measurement error, I have measured it with my good Latron RLC meter,

275 mH is too low and the no-load current would be too high, probably saturating the core. You need to review your measurements.
Yes.

XL = 2πLF = 2*3.1416*50 = 314Ω
I = V/X = 220/314 = 0.7A.

0.7A is far too higher magnetising current for a 200VA mains transformer. If your measurement is correct, then the transformer is bad. It most likely has a multiple shorted turns and should be scrapped.

#### soldar

• Super Contributor
• Posts: 2002
• Country:
##### Re: 20W 50Hz isolation transformer
« Reply #15 on: April 27, 2019, 07:10:19 pm »
I get 2*3.14*50*0.275=86.4

I = 230 / 86.4 = 2.66
All my posts are made with 100% recycled electrons and bare traces of grey matter.

#### Zero999

• Super Contributor
• Posts: 12730
• Country:
• Hero999
##### Re: 20W 50Hz isolation transformer
« Reply #16 on: April 27, 2019, 10:28:51 pm »
I get 2*3.14*50*0.275=86.4

I = 230 / 86.4 = 2.66
You're right. I missed the inductance off. The calculation I posted was for one Henry, not 275mH.

Smf