UC1843 PWM dc-dc up converter

Bill,
thank you for your interest in TI. Can you confirm your requirements? Forward converters act as bucks so must drop output voltage relative to input. Do you need a boost or flyback here?

You're feedback is noted. We are upgrading some datasheets and this is one so you should see some more DS info in 4-5 weeks.

If you clarify your requirements though I can try to find you a reference design. Or you can search here:www.ti.com/.../reference_design.page

Any reason you are looking at this part in particular ?

Regards
John

John,

Thanks for your initial response.  The full requirements are as follows:

1. Must provide ~40watts 1st version, ~80w for the 2nd rev.

2. Input voltage range from 22 - 38v.

3. Must provide ~28v output @ 40w/80w respectively

4. Needs 2% or better output regulation.

5. Needs to be reasonably high power efficiency  at least 85% or better.

6. Must be able to withstand radiation up to 30Krad.

7. Cannot use opto-isolators (can't survive radiation).

9. I have to have the 1st version design completed in about 4 weeks, so time is of the essence.

This particular PWM can operate in current mode and has duty-cycle capability from 0~100%, which allows it to operate in forward converter mode.  Flyback mode is OK as long as efficiency is not sacrificed.  However, typically forward converters are more efficient.  The most helpful items you could provide would be:

1. Full data sheet if it exists.

2. Schematics of a previous design that resembles this application as closely as possible.

3. Schematics of any evaluation board that exists now or in the past.

4. Any other suggestions on any other RH complimentary or companion parts.

5. Any thing else you can think of.

Thanks for the help,

Bill

Bill,

thanks for that info. I'll have our Hirel team comment here.

Regards,

John

Bill,

1. Can you highlight what is the application. What is being powered by 28V output.
2. Is this an Intermediate bus converter and you will stepping down 28V output further to supply other peripheral devices.
3. If so, why is 2% regulation critical?
4. Is isolation required and if so what isolation voltage must the design meet.
5. Output ripple requirements
6. transient requirements / step load/ slew rate
6. Operating temperature range
7. size constrain if any
8. Do you have discreet components i.e. MOSFETS etc in your AVL that has been qualified for radiation performance.

Ramesh,
Thank you for your response and questions. Answers:
1. RF PAs
2. & 3. The 28v must supply 90% of the total load directly, hence the regulation requirement. There are also other secondary voltages (regulators) required, but they are low power. There is need for a buck regulator that creates ~6v off of the 28v supply, then sub-divided after that with LDOs. Total power required from these voltages is ~6w.
4. 1000v isolation.
5. Post output-filter requirement is 200mv ripple at a given continuous load (3A max for final version).
6. -55 to 125C, 150C is even better.
7. Needs to fit on to 2x3" pcb, with parts on top/bottom, SMD and through-hole allowed.
8. I have located mosfet driver replacements that can withstand the radiation requirements.

-- Would the UC1846 controller be better with regard to any of the requirements I've listed ?
-- Is there perhaps a TINA sample-schematic utilizing the UC1843 or UC1846 that you could send or refer me to ?

Thanks,
Bill

Bill,

   See reference design using UC1825  and application note highlighted U110.  Design showcases push-pull topology with no isolation.  However if the IC is referenced  on the secondary side with its own bias supply.  Thus using a pulse x-fmr one can drive the primary side.

http://www.ti.com/product/UC1825/technicaldocuments

Same idea can be used with forward converter.

UC1901 isolated feedback generator would an alternate option,  however this part is not rad hard.   If the design opportunity justifies making process changes then that can be a viable option as well as it has tight reference regulation +/-1% at 25C  and +/-2% over full temp range.

Bill,

In order to get tight output regulation you can also post regulate the secondary output voltage.

Hi Bill,

Just an observation; with your input voltage range of up to 38V, you are going to need a separate boot-strap supply for the UC1843A as the maximum rated supply voltage is only 30V.

Given your power, space, isolation voltage, and time frame requirements, I would almost think you need to look into something off-the-shelf.  I don't want to steer you away from TI as they make great stuff, but a 100W half-brick module from a company like Interpoint or VPT might better suit your needs.  Given your time frame, the effort that would be required to fully design and test a converter to meet your specifications would be monumental.  Not saying it can't be done, but it will be extremely challenging.

Shawn