UCC28950: PMP8740 main transformer design

Hello Nidheesh

Given that the transformer core is the one you specified then the primary turns will have gone up by a factor of 4.6^2/1.8^2 which is about a 6 fold increase !.
I'd double check the core specifications - (especially the effective cross sectional area and initial permeability)

You should also check that the turns ratio is still correct - the easiest way is to measure the primary inductance and then the secondary inductance and then (Np/Ns)^2 = Lp/Ls

You should check the wire gauge that is being used - if there are really 6 times more turns then the wire will have to be thinner in order to fit in all the turns. This will increase the winding resistance significantly causing resistive losses.

I think you need to correct the transformer before you can make much more progress.

Regards
Colin

Hi Colin,

Thank you for your response.

The thing is that I want to change the voltage level. So I was using a turns ratio of 5 to match it with a 48V system.

But doing with PQ5050 core only.

So while calculations I needed around 1.8-2.4mH primary magnetizing inductance. 

So while winding 19-20 primary turns can give this inductance. But because of some mistake, I told my manufacturer to wind 25 primary turns and 5 secondary turns(still the turns ratio is maintained as 5).

This lead to the higher primary magnetizing inductance of 4.6mH.

So I want to know that this over-design of the magnetizing inductance can create any issue?

(Because some failure issues noticed after changing the transformer. the previous transformer was not capable of 48V but it was having 2.4mH inductance only, it was working fine. When I was trying upgrade voltage I got stuck with this issue)

Regards.

Nidheesh

Hello Nidheesh

I clearly remember answering this post but for some reason my answer doesn't appear in the thread - I'd really appreciate if you could check and confirm whether or not you received a reply.

Just in case you did not: here is a summary.

Excess Magnetizing inductance won't cause any real problems however the amount of energy available to drive the ZVS transitions will be less. Also, the contribution of the magnetizing current to the slope compensation ramp will be less so you might need to increase the slope compensation (RSUM) to compensate for this.

More turns normally means more resistance so higher copper losses. This will be offset by lower core losses. The nett effect of this may be to increase or decrease total losses.

Do please let me know whether my earlier answer reached you or not.

Here's some material on magnetizing current and ZVS - please let me know if the link does not work.

/cfs-file/__key/communityserver-discussions-components-files/196/7587.ZVS_5F00_SubSet.docx

Regards
Colin