UCC28950: work without current feedback

Vladimir Belyavsky

Part Number: UCC28950
Other Parts Discussed in Thread: UCC28951

Hi!
Some time ago I've been asking some questions about UCC28950.
Now I get real converter hardware for first run and testing.

Basic parameters are:
Full power 5kW.
Input 18-36V.
Output 400V (making test on 100V only for safety by changing feedback gain).
Frequency 50kHz.
Bridge is made with 16 MOSFETs by 4 MOSFETs in parallel.
There are 3300uF and 700uH output filter.

I have some doubt about correct working of current sensor. It's LEM505 with inverting OpAmp for scaling and rectifing as in TI Analog Engineer's cookbook, p. 70 ;)
I tried to start the converter without current feedback, but with only 50W load it gets unstable work with noise, big current transients and huge voltage spikes on MOSFETs.
Is it possible to run UCC28950 with voltage feedback only?

Now I want to test power capabilites and make some other tests. I don't need fast reaction on transients and full functionality.

P.S. I tried to run converter without any feedback - it's works correctlyff with 100% duty.

over 5 years ago

0 Peter Meaney over 5 years ago

TI__Mastermind 31580 points

Hi Vladamir,

I have asked one of my colleagues to respond to your post he has more experience with this type of cirucit

Is the LEM505 used for sensing the primary side switching current, does it give reasonable frequency response to accurately track the current wavefom?

Regards

Peter

0 Colin Gillmor over 5 years ago

TI__Mastermind 25570 points

Hi Vladimir

1/ Yes, it's possible to run the controller without feedback - it will operate at Dmax so you need to make sure you don't overstress the power stage. There will also be no current limiting. Below are two links, the first to a circuit that will run the UCC28951 in an open loop - although you seem to have been able to do this already and the second is a general purpose approach to debugging UCC28950 power stages

2/ I wasn't able to find the information in the Analog Engineers cookbook, nor anything about hall sensors - could you cut and paste an image of the page please ? I was also not able to find any information about the LEM505.

Can you post that part of the schematic here and I'll review it ?

3/ I've used Hall sensors in the past and they work well when used correctly. Their frequency bandwidth is the major characteristic to watch out for. They also are sensitive to the Earth's magnetic field so their output varies as the product is rotated - I found this out the hard way !.

4/ if there is excessive voltage spiking - and you will be able to see this if you follow the debug procedure then the most likely cause is that there is excessive unclamped inductance in the layout of the power stage.

//cfs-file/__key/communityserver-discussions-components-files/196/Open-loop.pdf

/cfs-file/__key/communityserver-discussions-components-files/196/5305.UCC28950_2C00_-UCC28951-Debug.docx

Please forward me the information above and I'll have a look at it.

Regards

Colin