This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
We're testing a flyback boost design using the UCC2803 controller. We used webench for an initial shot at it, but after 2 weeks of troubleshooting, are still struggling to get anywhere near the desired result.
Vin = 28VDC
Vout = 300VDC @ 5mA
fs = 300kHz (have also tried 100kHz)
We've tuned in snubbers across the switch, the D in the RCD clamp, The output D on the secondary... We've played wound several transformers with different turns ratios, higher primary turns than originally expected & even one with a segmented secondary architecture... We've played with a few different core gaps & an ungapped core set with an added padding inductor in series to match the expected leakage inductance given by webench. We even tried converting to the UCC2801, adjusting values & winding a new, higher turns ratio transformer to go with it.
The main issue is that we're not getting full duty cycle... according to Webench, we should be close to 60%. Does that ring a bell to anyone? I can provide many more details but don't want to get too far from the key observation right off the bat.
Hello Michael,
Sorry to hear about your design not working.
I used Webench with 24-32Vin, 300Vout, 100mAout using Flyback topology with external switch. I saw 1 design with the UCC2803-Q1 (click on schematic below). Did you get something similar? This design is recommending a custom transformer, which seems to be your biggest "unknown".
If this is not what your found, can you tell me exactly what you entered into Webench?
Did you choose the Flyback topology for a particular reason?
We have a very simple Boost reference design that you could get working with a high degree of confidence: PMP8621. This boost design is for 18-30Vin, 300Vout at 10mA. It seems to fit your requirements perfectly and is much simpler than the Flyback.
Regards,
Eric
Hi Eric,
I did 27-29VDCin, 300VDCout, 70C (temp), isolated, 5mAout. The chip I picked was UCC2803 as I'm using it for another multiple output flyback that needs isolation... technically, the 300VDC doesn't have to have isolation, so your reference design is quite interesting. By chance is there a TI TINA simulation for the PMP8621 design?
I would like to stick with the UCC2803 if possible for as wide of a temperature range as possible (the EP version).
Thanks,
Mike
Hi Mike,
There are a few TINA simulations for the TPS40210. Scroll down to the TINA Startup Reference Design shown below.
I think I would start with the Startup Reference design and re-configure it for the PMP8621 design.
First, I would run the Startup Reference Design without making any changes. Then, I'd re-configure it for the PMP8621 reference design.
By the way, I found a 3 year old E2E post where a user wanted to simulate the PMP8621 but initially failed. According to the last post, the model looks like it was updated, so I think it should work for you. Also, make sure you are running the latest version of TINA.
Regards,
Eric
Will do. My boss wanted me to ask if you guys have a physical sample of the PMP8621 reference design... like a eval kit prototype we could borrow. Any chance that's possible?
Hi Michael,
The reference designs are usually created by our Power Design Services team. I will ask if they have anything available and get back to you. Usually they create 1 or 2 units, document them, and provide them to a customer. This particular design was created back in 2013 so I'm not optimistic they will still have any units. On the other hand we assemble evaluation modules (EVMs) for products and these units are always replenished. If you order an EVM for the TPS40210 it may be something you can modify into the PMP8621 design.
Regards,
Eric