Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

High value NTC hampers Offline flyback startup

cupoftea

Advanced Member level 6
Advanced Member level 6
Joined
Jun 13, 2021
Messages
3,057
Helped
62
Reputation
124
Reaction score
139
Trophy points
63
Activity points
15,948
Hi,
Customer has sent us a 24V,1A offline flyback (90-265vac) to re-do
with a different controller (theirs went obsolete).
Its noteable that the FSGM0565 that they had been using has
a current limit of some 3A...so in effect they had no
realistic overload protection, but just short cct output protection.
(3A is well above their transformer's peak current capability, which is 300mT at 1.25A)

FSGM0565

Anyway, they are also using a 50 Ohm NTC...

B57236S0500M00 (NTC)

This has a resistance of 400 Ohms at minus 25 degrees C. (!!!)

The customer is reluctant to tell us what is the lowest ambient temperature where this flyback may operate.

Our proposed alternative controller has a startup voltage of 110VDC (ie the rectified mains, as sensed by the controller
must be at >110Vdc for at least some time within any 60ms time period.)

The thing is, that this wont be possible at minus 25degC, as the 400 Ohm NTC in the line wont allow the HVDC bus
to rise above 110Vdc for long enough. They insist on using the same 50R NTC. (they also insist on us using their same
flyback transformer which is E25/13/7 and ns/np = 19/84)

Anyway, would you agree that start-up testing at minus 25degc is going to be essential for this case? Also, are there any cheap places where
you can go and use a thermal chamber in UK? (sorry to speak of UK alone, please also tell for other countries as it
may be useful for people there).

This job has a sting in the tail IMHO, and this is the low temperature startup. We did wonder why they were using a FSGM0565 with
a ridiculously high primary current limit (3A) and a ridiculously low startup voltage of 28V. We reckon this is deliberate due to some customers having low ambeint temperatures.
_______________________ _________________________
Unfortunately, there are no offtheshelf offline flyback controllers which have all of ...
1...no UVLO
2...Output short circuit protection
3...No overload protection

...and this is what we need.
We cant use UCC38C42 because it doesnt have output short cct protection, and we have no room on the PCB to do it with discrete circuitry.
 
Last edited:
Hi,
The thing is, that this wont be possible at minus 25degC, as the 400 Ohm NTC in the line wont allow the HVDC bus
is this your assumption .. or really experienced on a real circuit?

****
You say it does not allow the DC bus to fully rise ... in other words .. there will be a significant voltage drop across the NTC.

Let´s assume worst case input voltage = 90V... and maybe there is a 20% voltage drop.
So let´s calculate with 18V of drop. This equals to a current of 45mA.
(I know this is not realistic because the current is not clean flat DC but recitified AC. So correctly we had to calculate the RMS voltage across the NTC .. but it would be higher than these 18V. So using 18V is still on the safe side)

OK. Now these 18V (treated as RMS) will cause a 0.81W of heat on the 400 Ohms.
Taking the 10mW/K from the datasheet ... 0.81W will cause a rise in heat of 81K or from -25°C to +56°C ...

--> The NTC will heat up from -25°C .. to a higher value. And thus automatically gets a "valid operation temperature". I see no problem.

Klaus
 
Thanks, but the NCP1239E is already seen to be not starting up at 90VAC when at room temperature.
(When we short out the 50 Ohm NTC it does however start up at 90VAC)
With the NTC in circuit, we have to raise the mains to 100VAC before it starts up.

This shows the effect of the 50 Ohm NTC, and we can only imagine, this will be much worse at lower ambeint temperatures when the NTC is much higher resistance....its 400 Ohms at -25degC.

As can be seen, the NCP1239E should run as long as its HV pin sees at least 101V during any 64ms period. (If during any 64ms period it doesnt "see" 101V or more, then it shuts down...and thats what we are seeing, repeated shutting down at 90VAC mains.)

NCP1239E (pg3 shows the UVLO voltages)
 
If it were me I'd experiment with resistor values in parallel across the thermistor. Say another 400Ω in parallel makes 200Ω overall and that ought to make voltage rise sufficiently at startup. Once the equipment is operating the thermistor heats up (per Klaus' insight post #2), its resistance drops and it takes the main burden.

thermal chamber

Dry ice (frozen CO2) has a temperature much colder than -20F. Food/drink vendors sometimes sell it cheap.
 
This shows the effect of the 50 Ohm NTC
If you want to do the investigation style ... then you have to measure the currents in the different modes of operation.
(at various voltages, at fail/no fail condition, at different temperatures)

Or you do it trial and error wise

Klaus
 

LaTeX Commands Quick-Menu:

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top