TPSF12C3QEVM: Design with the TPSF and Nanocrystalline CMC from Magnetec

Thomas wentzlik

Part Number: TPSF12C3QEVM
Other Parts Discussed in Thread: TPSF12C1, TPSF12C3

My CS1675615:

yes I did start the chat and after a short time it did not work on my side.

The reason for contacting TI is to get a technical feedback how this TPSF12Cx can be used with a Nanocrystalline CMC (Magnetec GmbH is a manufacture of this Nanocrystalline material).
In this case I am interested in samples demoboards and the question how this concept can be used for high power applications >>50kW

The sample and demo board question finally I found a answer on the Webpage, but for the higher power application I am still thinking.

over 2 years ago

0 Timothy Hegarty over 2 years ago

TI__Guru 74162 points

Hi Thomas,

Yes, a 50kW application is feasible. I assume this is a 3-phase application. If you have an existing filter schematic, we can recommend a suitable AEF circuit. The two CM chokes in a two-stage passive design can typically be reduced by a factor of 5 in an equivalent active design.

Regards,

Tim

0 Thomas wentzlik over 2 years ago in reply to Timothy Hegarty

Prodigy 30 points

Hi Tim

thanks for the feedback, we will take a look into the design and get back to you for support.

Do you have a CMC demoboard for the TPSF12C3 or the TPSF12C1?

Regards

Thomas

0 Thomas wentzlik over 2 years ago

Prodigy 30 points

Hi Tim

we have a concept design with

Three Phase with Neutral

Sorce Impedance 1uH 5mOhm

2 x CMC 13mH and DM=800nH

Grid Side X 5nF
Grid Side Y 2.2nF 1Ohm 10nH

Load Side X 500nF
Load Side Y 2.2nF 1Ohm 10nH

Can this be used for filter schematic info?

Regards

Thomas

0 Timothy Hegarty over 2 years ago in reply to Thomas wentzlik

TI__Guru 74162 points

Hi Thomas,

I recommend completing the quickstart calculator. Here is the file:

TPSF12Cx quickstart calculator Rev A2.xlsm

Regards,

Tim

0 Thomas wentzlik over 2 years ago in reply to Timothy Hegarty

Prodigy 30 points

MAGNETEC_DRAFT_TPSF12Cx quickstart calculator Rev A2.xlsm

Hi Tim

can you take a look into the file uploaded file? With this values I used a different filter values and you can see some data from the CMC (start with 10kHz) and I adjust the values Step1 - 3. The other Steps are default values.

What I was missing is the X capacitor values CX1..3 CX2..3

Thanks

Thomas

0 Thomas wentzlik over 2 years ago in reply to Thomas wentzlik

Prodigy 30 points

Hi Tim

how are you? Could you take a look in the TPSF12C.... attached. Can this be used and improved with the active filter IC to optimize the design?

Thanks

Thomas

0 Thomas wentzlik over 2 years ago in reply to Timothy Hegarty

Prodigy 30 points

Hi Tim

how are you? Could you take a look in the TPSF12C.... attached. Can this be used and improved with the active filter IC to optimize the design?

Thanks

Thomas

0 Timothy Hegarty over 2 years ago in reply to Thomas wentzlik

TI__Guru 74162 points

Hi Thomas, I'll take a look now and get back to you asap.

Tim

0 Timothy Hegarty over 2 years ago in reply to Timothy Hegarty

TI__Guru 74162 points

Thomas, is it 9.7nF for each grid-side and regulator-side Y-cap (i.e. 2 x 9.7nF on each side)? Also, the CM inductance seems low here, hence the resonant frequency of the chokes and inject cap appears at a high frequency ~100kHz.

Try increasing the inject cap to 10nF or 22nF if you can't increase the CM inductance - that will reduce the resonant frequency and prevent the phase from reaching -180 (which indicates instability).

Tim

0 Thomas wentzlik over 2 years ago in reply to Timothy Hegarty

Prodigy 30 points

Hi Tim thanks for the feedback, yes we have to change the Y-cap values to a better value and this needs to be checked with the customer. The idea was more can this TI active filter IC support in this area? Maybe also to reduce the CM inductance.?

Thomas

0 Timothy Hegarty over 2 years ago in reply to Thomas wentzlik

TI__Guru 74162 points

Hi Thomas,

The AEF IC should help to reduce CM inductance. It would be good to know the EMI source amplitude to determine if the filtering as shown is adequate.

Tim

0 Thomas wentzlik over 2 years ago in reply to Timothy Hegarty

Prodigy 30 points

Hi Tim we have a couple measurements and this shows a peak at 20MHz, but we see also some noise in the <1MHz area and maybe this TI AEF can support the design in this frequency area. We need to build a filter prototype and do testing, and for this a EVM would be helpful. In the Excel simulation file of the TPSF I see the step 6 network parameter have an significant effect on the damping curve.

The high current is critical to keep the CMC small, or increase the number of turns, because we want to use a high permeability.

Does this info support your question?

Thanks

Thomas

Power management

Power management forum

TPSF12C3QEVM: Design with the TPSF and Nanocrystalline CMC from Magnetec