Hi guys.
hi guys starting new thread, i got my answers from the previous ones.

i have recently acquired a microwave transformer, now i want to rewind the secondary WITHOUT touching the primary(but if i must i will), so it goes without saying then i do not know the amount of turns on the primary.... HOWEVER my targeted output is ~36V.... what is the best way going in calculating it??,,, can i just take my mains 220v/36v = 6.11 then check either the primary's inductance or resistance, and then divide THAT by 6.11 to see my targeted resistance or inductance and then just wind the secondary until i achieved the needed amount?

ALSO in testing that i have got my targeted output, will it be okay to put a 1M ohm resistor across the secondary's points and measure the voltage across it in AC mode? in my head 36V/1Mohm= very small current, so nothing should over heat and pop. unless you guys know something bout AC that will make it go pop. hahahahaha

Hi,

The main thing about these is the excitation current and as others have mentioned, the shunts, and the all-important experimentation. Simple experiments are needed to be able to get the specs right for use in a DC power supply.

The excitation current will be high, and the shunts have to be removed so you don't get too much voltage sag. The tradeoff in removing the shunts is you need more filtering of the rectified output once you are done with the conversion.
Since the excitation current will be high that means it could easily overheat. Adding more primary turns will help with that.

After you remove the secondary you can run up the primary and check the excitation current. You want this down to less than 10 percent of the full load current. The full load current will depend on the wire size of the primary, and the secondary wire size comes from that also. The voltage ratio is Vout/Vin, and the current ratio is Iout/Iin. The voltage ratio determines the output voltage and the current ratio determines the max current you can get from the output coil. The secondary turns must also fit on core which means you have to figure in the window area too.

Once you add the extra primary turns, then you can experiment by testing the output voltage with a test secondary coil. If you wind a coil with 10 turns and you get 5 volts, then you know the ratio is 0.5 volts per turn. You can then figure out how many turns you need, and the max wire diameter will come from the size of the window area you have for the secondary turns. It is also wise to load the turns with the full load current and measure the voltage a second time so you know the loading factor. If we had 10 volts output with no load and 9 volts out with full load, then the loading factor would be 90 percent or 0.90, and that would relate to the wire diameter. If you have enough window area you can go up in wire diameter and thus get a better loading factor, maybe 0.95, but if the window area is even more limited you may have to go down with the wire diameter and that would of course make the loading factor worse like maybe 0.85 or so, and if you reach a point where the loading factor is too bad you may have to work that into the DC regulation scheme so you eventually can get a constant DC output from the entire power supply.

It's also good if you own a variac so you can do lower voltage testing with lower voltages on the primary as you do some experiments. That also helps if you want a variable raw output from the transformer once it is rewound. That way you get a lot of variability plus isolation from the mains line.

 

POst #31 mentions EXCITATION CURRENT as well as adding primary turns. What is the issue about the excitation current, prior to adding a different secondary?? And why would I need to add primary turns???
I will need to go back and see what I have missed.

 

POst #31 mentions EXCITATION CURRENT as well as adding primary turns. What is the issue about the excitation current, prior to adding a different secondary?? And why would I need to add primary turns???

EXCITATION CURRENT can be high due to magnetic saturation in a µW transformer designed for intermittent use, and causes transformer heat build-up to likely damaging levels in continuous use, even with no output load.
Adding primary windings reduces that saturation and thus the current, to reduce the idle heat.

 

I had not thought about saturation being caused by a reduced load. A very good point to watch out for.
I may have enough room for more of what looks like about #18 copper wire. My secondary will be two strands of #12 wire. Good for 50 amps short term., 40 amps steady state. BUT THEN i WILL NEED 4 BIG 100 AMP DIODES IN THE BRIDGE.

 

POst #31 mentions EXCITATION CURRENT as well as adding primary turns. What is the issue about the excitation current, prior to adding a different secondary?? And why would I need to add primary turns???
I will need to go back and see what I have missed.

Hi,

As has been mentioned already, a high excitation current means the core is nearing saturation, and that means the current could be much higher than the normal operation would be if it did not get near the limit.

Since the primary turns is strongly related to the degree of saturation, we start to talk about adding primary turns. That's because the more primary turns you add the lower the excitation current and the farther away from saturation the core gets.
So more turns = less saturation and less excitation current, and also provides more of a margin for the high line mains power line spec which is 10 percent higher than nominal (132vac on a 120vac line).

The other point is that when they make transformers for microwave ovens, they tend to use less primary turns than what would keep it further away from saturation. That's to keep cost and weight down mostly, and also because if the transformer is not run for too long at one time it won't get as hot.

Now the degree of saturation right out of the factory you don't know until you measure the current into the transformer with no load. Once you do that you can decide if you want to add more primary turns (using the same wire size as the existing primary wire size).

 

Hi,

As has been mentioned already, a high excitation current means the core is nearing saturation, and that means the current could be much higher than the normal operation would be if it did not get near the limit.

Since the primary turns is strongly related to the degree of saturation, we start to talk about adding primary turns. That's because the more primary turns you add the lower the excitation current and the farther away from saturation the core gets.
So more turns = less saturation and less excitation current, and also provides more of a margin for the high line mains power line spec which is 10 percent higher than nominal (132vac on a 120vac line).

The other point is that when they make transformers for microwave ovens, they tend to use less primary turns than what would keep it further away from saturation. That's to keep cost and weight down mostly, and also because if the transformer is not run for too long at one time it won't get as hot.

Now the degree of saturation right out of the factory you don't know until you measure the current into the transformer with no load. Once you do that you can decide if you want to add more primary turns (using the same wire size as the existing primary wire size).

Many Thanks, Me.Al! A very good explanation.

Now a caution for the original poster: I see one end of the secondary tied to a ground lug on the transformer frame. The other end will be at the full AC high voltage, You need to know where that connection is located on the transformer, and avoid contact with it.

It is easily a thousand volts at a hundred milliamps. A VERY LETHAL power level!!!

 

You help very much. thank you. ok i have updated my simulations, but it seems to have errors. i guess trial and errors in real life will help me tune better. HOWEVER one can use the simulation to prepare.
So in real life:
1- i will check the magnetron heater output voltage and then see the turn, and will know the turn per volt i need. (as was mentioned)
2- I will wind it to the RMS voltage, so if i want 35-36Vdc i need 36v/1.141 = 25VAC on secondary... (trial and error)
3- i am looking for MAX power output, that mean i must rate my wires and diodes to the amperage, max input with before 10amp fuse blows is 2300W(south african power is 230V 50Hz),,, so assuming 100% efficiency (yes i know it's not, but preparing for it is smart) then i nee 2300/36v=63 amps,,,, thus to rectify(i do not have 100amp FBR i need for each of the 4 diodes double 30amp)....

then here comes the question, HOW DO I KNOW what capacitor to use for smoothing. online it gives this formula
View attachment 360714
using that i see i need 10mF.... but the ripples on load are still HUGE
View attachment 360715
the output power is what i want, on max, but the ripple and input power is unreal, why am i drawing 15amps that its should be around 5, or at least under 10 amp. as 230V at 1200W should give about 6amps. what you put in is what u should get out, minus losses..im gonna kill the breaker. in a second
also my transformer primary inductance must also be set to 1H, if i set it to the actual value measured 13mH then i draw 110Amps from mains in the simulation, that 23000W wtf.... so i do not know if that is real.
View attachment 360716
i also needed to increase output to high volts output on the double diodes in the simulation. in real life it will be trial and error again.

i also added an inductor before the capacitor but that drops my output power by a lot, no matter what the value of the inductor, also volts drop to unusable as my load requires MINIMUM 24Vdc
View attachment 360717

is my filtering correct? how can i make it better with required output? max power 36Vdc at 20Amp and up?
using the formulate above to calculate the capacitor gives weird results.

no one answered this. or even tried.

 

First, my trick for learning to find turns per volt is in post #6.
As for filtering out the ripple, if you have an amplifier circuit schematic, it may include the power supply, and mention capacitor values. It may also mention the required current for the amplifier rated output.
For capacitor values, some of us choose the highest value at the rated voltage that will fit, or that we can afford.
One more suggestion that will help with the filtering arrangement::
Go to the website: "Schematics For Free", and then to the audio section, and amplifiers. You can see what is done in similar amplifiers, and see what values of capacitors are used in good quality equipment. Probably no formulas, but circuits of what has worked for others. Studying success is a useful way to learn. Sort of cheating a bit, but seeing what works can be very useful, even if you are not told how they got the answers. AND IT IS FREEE!!!

 

no one answered this. or even tried.

Hi,

20 amps is a pretty high load current. That most likely would require some fairly large capacitor values. You also have to pay attention to the ripple current or the cap(s) can overheat. That means parallel caps rather than just one large value cap.

I have pictures of the one I modified a long time ago I'll see if I can find them. It was not easy getting that secondary coil off using a hacksaw and center punch and hammer to push out the coil wires. So far I did not wind a secondary on it yet as I got involved in other things right after that
It was my intent to make a high current DC power supply, maybe 50 amps or something.

 

20amps at 1000 watts is 50 volts, at 100% efficiency. Adequate for wanting 36 volts.