Electronics > Beginners

Basics on winding your own transformers

(1/4) > >>

Dan123456:
Hey all  :)

Hopefully a quick one.

I am looking to make a simple single phase step up transformer that turns 12VAC, 0.3A to 90VAC (I don’t really care about output amps too much but my understanding is this would be 0.04A) to power a neon bulb.

Why?… well simply just to play with them cos neon look cool!

I was thinking of doing this with 1.25mm copper wire on the primary and either 0.8mm or 0.3mm on the secondary with the primary and secondary wrapped around opposite sides of a 35mm ferrite “donut”. I know this wire gauges are way thicker than what I theoretically need but that is simply just to make it easy for me to build.

I have 2 questions though,

1. Does the wire gauge make a difference to anything other than the just max current it can take? I was thinking that 50 turns of 0.8mm wire would have a lot more surface area than 50 turns of 0.3mm so surely that must effect something yeah?

2. Following on from that logic, is there a rule of thumb / way to work out how many turns of the primary you should do? I understand I would need a turn ratio of 7.5 in this situation but can I just do any old number of turns on the primary then 7.5 times that on the secondary? Again, 1 turn would have less surface area compared to 10 etc. so I assume this would effect something right?

Thanks so much in advance :)

srb1954:

--- Quote from: Dan123456 on October 02, 2023, 09:52:58 am ---Hey all  :)

Hopefully a quick one.

I am looking to make a simple single phase step up transformer that turns 12VAC, 0.3A to 90VAC (I don’t really care about output amps too much but my understanding is this would be 0.04A) to power a neon bulb.

Why?… well simply just to play with them cos neon look cool!

I was thinking of doing this with 1.25mm copper wire on the primary and either 0.8mm or 0.3mm on the secondary with the primary and secondary wrapped around opposite sides of a 35mm ferrite “donut”. I know this wire gauges are way thicker than what I theoretically need but that is simply just to make it easy for me to build.

I have 2 questions though,

1. Does the wire gauge make a difference to anything other than the just max current it can take? I was thinking that 50 turns of 0.8mm wire would have a lot more surface area than 50 turns of 0.3mm so surely that must effect something yeah?

2. Following on from that logic, is there a rule of thumb / way to work out how many turns of the primary you should do? I understand I would need a turn ratio of 7.5 in this situation but can I just do any old number of turns on the primary then 7.5 times that on the secondary? Again, 1 turn would have less surface area compared to 10 etc. so I assume this would effect something right?

--- End quote ---
You need to have sufficient turns in each transformer winding to ensure that the peak flux density doesn't get to high. Having too few turns will saturate the core material, lowering the magnetising inductance of the transformer and causing excessive current to be drawn from your power source.

The governing equation is:

N = Vrms/(4.44 x f x B x A)

where Vrms is the winding voltage, f is the supply frequency,  A is the effective cross-sectional area of the core and B is the maximum allowable flux density, typically 1.2 Teslas for laminated iron cores and 0.3T for ferrite cores.

MrAl:

--- Quote from: Dan123456 on October 02, 2023, 09:52:58 am ---Hey all  :)

Hopefully a quick one.

I am looking to make a simple single phase step up transformer that turns 12VAC, 0.3A to 90VAC (I don’t really care about output amps too much but my understanding is this would be 0.04A) to power a neon bulb.

Why?… well simply just to play with them cos neon look cool!

I was thinking of doing this with 1.25mm copper wire on the primary and either 0.8mm or 0.3mm on the secondary with the primary and secondary wrapped around opposite sides of a 35mm ferrite “donut”. I know this wire gauges are way thicker than what I theoretically need but that is simply just to make it easy for me to build.

I have 2 questions though,

1. Does the wire gauge make a difference to anything other than the just max current it can take? I was thinking that 50 turns of 0.8mm wire would have a lot more surface area than 50 turns of 0.3mm so surely that must effect something yeah?

2. Following on from that logic, is there a rule of thumb / way to work out how many turns of the primary you should do? I understand I would need a turn ratio of 7.5 in this situation but can I just do any old number of turns on the primary then 7.5 times that on the secondary? Again, 1 turn would have less surface area compared to 10 etc. so I assume this would effect something right?

Thanks so much in advance :)

--- End quote ---

Hello there,

Well, there are a number of things to think about.

The first has already been mentioned in the previous post.  It's sometime called the "transformer equation" but there are a number of these equations that's just one of them.  It really relates the primary inductance to the turns, frequency, applied voltage, core area, so that the primary current does not go too high can cause the metal core to become ineffective.

The second would be the type of core material.  For line operated transformers the usual choice is silicon steel laminations, often in the form of an "E" and an "I".
You can find this kind of construction in old microwave oven transformers.  In fact, many people have cut off the secondary and provided their own secondary, although i also recommend adding some primary turns too because the mic oven transformers are not made as well as a 24/7 run regular line transformer.
It's a bit hard to get the scondary off, but you can do it.

The third is insulation.  You have to have good insulation between the primary and secondary or else you subject the end user (yourself) to a dangerous shock risk.  In fact, higher class transformers do not put the secondary right on top of the primary, they put the windings side by side with a spacer between.  That helps to eliminate the crushing problem where the transformer is accidentally crushed and the secondary gets pushed into the primary and thus the secondary ends up having the line potential on it as well as the primary.  You want to keep the secondary isolated as well as possible.

The fourth is wire sizes.  The current is inversely proportional to the voltage, so you can size the wire for the secondary based on the wire size of the primary.  The key here is not the surface area of the wire as much as the cross sectional area.  Basically if the input voltage was 10v and the output was 100v and the input current was 1 amp, then the output current would be 0.1 amp. This 10 fold ratio means you can use wire with one-tenth the area of the primary for the secondary.
There is also the loading factor that kicks in here too though.  You have to consider how much the secondary voltage will sag under full load.  That means you may have to go slightly higher for the secondary voltage and use a slightly larger area for this wire.  For example, instead of going to one-tenth the area for the secondary, maybe just one-eighth or something like that.  You figure how the loading based on the total resistance of the primary and total resistance of the secondary, along with the turns ratio.  The primary resistance reflects to the secondary as the square of the turns ratio.  That means once you reflect this resistance to the secondary, you add that to the secondary resistance to get the full loading series resistance on the secondary.

There are a few other things to consider but these are the most important for line operated transformers.
For example, you have to include some sort of fusing, probably on the primary and on the secondary too for the most safe way to do it.

Zero999:
Don't bother making your own. Use a mains transformer, in reverse.

You need a turns ratio of 1:7.5

A mains transformer with a 120V primary and 15V secondary, run in reverse should give just over 90V out, but in reality it'll be slightly less, as the 15V winding will have a slightly highe voltage, to compensate for loading.

Note, your neon lamp will need a resistor to limit the current, as it has a sharply negative voltage/current curve.

Dan123456:
Ah! Thank you so much!

This is exactly what I was looking for  :D

Just wondering, what units should I be using?

I tried googleing the formula and it seems like people are using all different units with different variations on the formula (such as Hz vs MHz and cm2 vs m2 vs inches2 or gauss rather than T).

Firstly, I worked out the effective cross-sectional area of my rings =“35x21x13” = 7mm walls X 13mm height = 91mm2.

The answer to the equation that looked the most “right” to me was N = 12VAC /(4.44 x 60Hz x 0.3T x 0.091m2 = 1.65 turns (this also used the units from a picture of a textbook I found - below).

Tried a few other units and got number ranging from virtually 0 to ~1700 turns and, I have no reference for what “right” looks like, I figured I’d just ask  :P

P.s. actually… is the 12VAC right or should that be the highest voltage out of the two (I.e. 90V or 90Vrms - I didn’t bother with the rms for the calcs at 12v as was negligible but it would make more of a difference at 90V)? I assumed it meant primary coil input voltage but just thought of that and, if that is the case, that gives me roughly 12 turns which sounds much more reasonable!

Thanks so much again!