help regarding the transformer design parameters from a company

Hello All. I ordered a transformer for 48V input and 330V output. I want to know whether the provided values of the primary and the secondary inductances are okay or not? My primary current can go upto 21A. And the switching frequency is 100kHz, 1000W.
Attached is the document for the specs from the factory.
Also I want to know that if the actual inductance falls short of the required inductance, then what would be the outcome?
Thankyou in advance.

Can you provide more information on the topology you are planning to use?

E-design said:

Can you provide more information on the topology you are planning to use?

Click to expand...

I am using the the Full-Bridge topology converter. With the full wave rectifier (Full bridge).

Do you have a Vin_min and Vin_max specification?

usama14 said:

Also I want to know that if the actual inductance falls short of the required inductance, then what would be the outcome?
Thankyou in advance.

Click to expand...

Smaller inductance admits greater current within a given time. You can expect to adjust switching frequency, to obtain desired performance.

Since I like to watch current bundles go round loops, here's a simple simulation of your transformer with 48V square waves.



From my (theoretical) experiments with transformer primary Henry values, it appears they can occupy a wide range and still perform efficiently.

At 100 kHz flux max density should be limited to 150 mT to keep core losses under control.

Working with Vin_min = 45 V and Vin_max = 48 V gives;

Np = 2
Ns =15
Inductance(p) = 21 uH
Peak Ip = 28.5 A
Core loss ≈ 11 W

It appears your transformer design will allow for Vin_min = 38 V and a low peak flux of 80 mT.
So it is a very conservative design that will run cool with only about 1 W of core losses.

Peak primary current will be about 29 A and average operating current 22.6 A.

BradTheRad, Wouldn't the secondary of 7.8 mH be the worst case for 100 KHZ, compared to the 95 uH primary?

- - - Updated - - -

E-design said:

At 100 kHz flux max density should be limited to 150 mT to keep core losses under control.

Working with Vin_min = 45 V and Vin_max = 48 V gives;

Np = 2
Ns =15
Inductance(p) = 21 uH
Peak Ip = 28.5 A
Core loss ≈ 11 W

It appears your transformer design will allow for Vin_min = 38 V and a low peak flux of 80 mT.
So it is a very conservative design that will run cool with only about 1 W of core losses.

Peak primary current will be about 29 A and average operating current 22.6 A.

Click to expand...

My numbers were similar to yours with a little less on the flux level. Yata chose a flux level of 815 Gauss. My guess was to lower the core loss to make up for the copper loss of primary and secondary copper thickness a little more than desired.

FlapJack said:

BradTheRad, Wouldn't the secondary of 7.8 mH be the worst case for 100 KHZ, compared to the 95 uH primary?

Click to expand...

Sorry, I don't know how to test for this. The simulator only displays transformer parameters for:
* primary value
* turns ratio
* coupling coefficient

Internally I suppose it makes use of the normal formula for deriving the secondary value, given the other parameters.

E-design said:

At 100 kHz flux max density should be limited to 150 mT to keep core losses under control.

Working with Vin_min = 45 V and Vin_max = 48 V gives;

Np = 2
Ns =15
Inductance(p) = 21 uH
Peak Ip = 28.5 A
Core loss ≈ 11 W

It appears your transformer design will allow for Vin_min = 38 V and a low peak flux of 80 mT.
So it is a very conservative design that will run cool with only about 1 W of core losses.

Peak primary current will be about 29 A and average operating current 22.6 A.

Click to expand...

So the design parameters are okay in your opinion then?

- - - Updated - - -

BradtheRad said:

Smaller inductance admits greater current within a given time. You can expect to adjust switching frequency, to obtain desired performance.

Since I like to watch current bundles go round loops, here's a simple simulation of your transformer with 48V square waves.



From my (theoretical) experiments with transformer primary Henry values, it appears they can occupy a wide range and still perform efficiently.

Click to expand...

I had a similar question though. Actually I made another transformer, but it has low primary inductance value due to the breakage of the core. And by changing the duty cycle value, Im not getting a significant change in the output voltage. Is it that the transformer is getting saturated or something? Or is it bcz of the inductance values?

usama14 said:

So the design parameters are okay in your opinion then?

Click to expand...

It should work just fine according to the specifications.

E-design said:

It should work just fine according to the specifications.

Click to expand...

Ok Thankyou.
And can you kindly answer my second question regarding the other transformer I made myself.

usama14 said:

I had a similar question though. Actually I made another transformer, but it has low primary inductance value due to the breakage of the core. And by changing the duty cycle value, Im not getting a significant change in the output voltage. Is it that the transformer is getting saturated or something? Or is it bcz of the inductance values?

Click to expand...

Let's see a theoretical comparison. Add a transformer with 1/2 the primary value, and another with twice the primary Henry value.



Differences are slight. Again, this is only theory.

Is it that the transformer is getting saturated or something?

Click to expand...

The lesser primary value has a rising waveform and it will draw greater current. If that peak draw goes above saturation level, then it would cause deteriorated output.

It is impossible to give a useful opinion without the actual transformer specifications.

E-design said:

It is impossible to give a useful opinion without the actual transformer specifications.

Click to expand...

The transformer I made had 4 turns primary and 36 turns on the secondary with the inductance values pretty low. I designed it for 100kHz operation.

- - - Updated - - -

BradtheRad said:

Let's see a theoretical comparison. Add a transformer with 1/2 the primary value, and another with twice the primary Henry value.



Differences are slight. Again, this is only theory.



The lesser primary value has a rising waveform and it will draw greater current. If that peak draw goes above saturation level, then it would cause deteriorated output.

Click to expand...

Deteriorated output? What does that mean?

I wrote:

If that peak draw goes above saturation level, then it would cause deteriorated output.

Click to expand...

usama14 said:

Deteriorated output? What does that mean?

Click to expand...

You can send very high A through the inductor, but the flux field only reaches a certain intensity (saturation level), and then does not rise further. Then as soon as you start to draw current from the inductor, the A you get is less than the previous A you fed it. Therefore you do not get back all the energy you put in. That is the deterioration I referred to.

BradtheRad said:

I wrote:





You can send very high A through the inductor, but the flux field only reaches a certain intensity (saturation level), and then does not rise further. Then as soon as you start to draw current from the inductor, the A you get is less than the previous A you fed it. Therefore you do not get back all the energy you put in. That is the deterioration I referred to.

Click to expand...

So does this relate to the non-changing output when the Duty Cycle is changed? Actually the output changes in stages of 5V when the duty cycle is changed by 5% for the 2k ohm resistor. But as soon as the load is increased, the output remains constant at 148V dc. That's where I'm all stuck right now. :/

usama14 said:

So does this relate to the non-changing output when the Duty Cycle is changed? Actually the output changes in stages of 5V when the duty cycle is changed by 5% for the 2k ohm resistor. But as soon as the load is increased, the output remains constant at 148V dc. That's where I'm all stuck right now. :/

Click to expand...

From what the other replies state, your transformer should operate okay and not saturate. I hope I did not appear to talk as though I was contradicting the other replies (or your factory specs).

There is a different problem which is likely to pop up, namely parasitic resistance. This can appear in several places, and it will hamper your power transfer. To cure it:

* Every switching device has to be biased sufficiently so it presents minimum resistance.

* Every wire connector must be attached securely. All contact surfaces need to be clean and shiny.

* Your power source needs to have very little internal resistance.

All this adds up to, is you need to make sure you can get 48V 21A going through your transformer.

BradtheRad said:

From what the other replies state, your transformer should operate okay and not saturate. I hope I did not appear to talk as though I was contradicting the other replies (or your factory specs).

There is a different problem which is likely to pop up, namely parasitic resistance. This can appear in several places, and it will hamper your power transfer. To cure it:

* Every switching device has to be biased sufficiently so it presents minimum resistance.

* Every wire connector must be attached securely. All contact surfaces need to be clean and shiny.

* Your power source needs to have very little internal resistance.

All this adds up to, is you need to make sure you can get 48V 21A going through your transformer.

Click to expand...

Hmm...I was testing my own transformer on 24V dc batteries for now(The one I made myself, not the factory version as it hasnt reached me as of now). And as I'm building a dc-dc converter with the H-bridge design, after the turns transformer secondary and the full bridge rectifier, I am getting the following results:

1) Duty cycle=85%
Load resistor=2k ohm
VpeakTopeak at the secondary=50.8 cross 10 = 467.36
Vdc(that should be the output)= 2Vm/pi =146V
Vdc practical value=143V Sounds GREAT. BUT..
But when I decrease the duty cycle, the output doesnt change a bit. Even uptill duty cycle of 60%. It rather starts to decrease at some points too.

2) But in case of lower load like R=820 ohm,
Duty cycle=85%, Vdc practical value=133V
Duty cycle=80%, Vdc practical value=124V
Duty cycle=75%, Vdc practical value=117V
Duty cycle=70%, Vdc practical value=112V
Duty cycle=65%, Vdc practical value=107V
Duty cycle=60%, Vdc practical value=103V
Which is pretty much acceptable.

But I cant understand a bit about this problem. In some cases the PEAK TO PEAK voltages observed at the secondary are not conforming to the formula of Vdc=2Vm/pi and the output seems higher than the theoretical values :/
My supervisor suggested that the transformer may be getting saturated and the core requires an air gap. Can you kindly explain these problems?

There is no information about the leakage inductance, at 100kHz, XL = 2.pi.F.L this inductive reactance limits the amount of power the transformer can let thru, as you load the Tx more and more the Vout will drop due to winding R and total leakage (Il x XL)... having a simple HV sec over a LV primary will give 4x the leakage than for splitting the HV sec and putting either side of the LV primary.
p.s. you can measure the Lleak by shorting the LV side (properly) and then measure the inductance from 16-17 to 22-23, this gives the total leakage referred to the HV side.

Easy peasy said:

There is no information about the leakage inductance, at 100kHz, XL = 2.pi.F.L this inductive reactance limits the amount of power the transformer can let thru, as you load the Tx more and more the Vout will drop due to winding R and total leakage (Il x XL)... having a simple HV sec over a LV primary will give 4x the leakage than for splitting the HV sec and putting either side of the LV primary.
p.s. you can measure the Lleak by shorting the LV side (properly) and then measure the inductance from 16-17 to 22-23, this gives the total leakage referred to the HV side.

Click to expand...

You mean to say that if I want to decrease the losses, I should make 32+32 center tap secondary rather than a single 36 one?