LT1172 3 V to 300 V converter issues

Question

I’m building a circuit around an LT1172 to deliver 0-300 V from a USB charged LiPo cell. I’ve included the circuit.

I’m using an EPCOS EFD20 former with E87 cores (0.3 mm gap).

If I just have one of the output windings the circuit works like a charm, with an arbitrary 200 turns on a single secondary I get a solid 140 V into a 45 K resistor which together with the 100 K feedback pot works out at just over 0.6 W and holds line regulation between 3 V and 4.2 V input and load regulation doesn’t budge, efficacy is 65 % which I think isn’t too shabby, it also adjusts from 300 V down to 1.25 V, with a 2.4 V Zener in series with the output I can get 0-300 V, I couldn’t ask for more.

Now the issue is when I wind the second secondary everything goes belly up. Both windings end up going down to 84 V, I can tweak these back up to 130 V however I loose regulation and when I connect the two stacked outputs together the total voltage collapses further and the input current drops. Originally I used an online flyback calculator and entered the following:

100 kHz

Eff 60%

Al 0.25

Vin 3 V

Vout1 150 V

Iout2 0.0022 A

Vout2 150 V

Iout2 0.0022 A

This resulted in:

Lp 19 uH

Np 8.74 (9)

Ns1 659

Ns2 659

The 9 primary turns didn’t seem to put enough energy into the core and even a single 659 turns of secondary just killed the circuit, so I wound 19 turns on the primary (48 uH) and took a punt on 200 turns for the secondary which worked perfectly as mentioned.

Obviously the high number of turns on the secondary is causing the issue but I seem to have hit a brick wall and can’t get anywhere near my goal of 300 V.

I’ve built a few of these 300 V circuits before but used a XP A03P-12 which are very expensive and need a 12 V supply as well as not being adjustable.

Anyone know what I’m doing wrong?

edit. The core in the picture is separated, this is not how it was being used, I found by separating the cores I could get another 10 V of output.

Answer 1

so I wound 19 turns on the primary (48uH) and took a punt on 200 turns for the secondary which worked perfectly as mentioned.

OK with 300 volts on the output, the flyback voltage on the primary will be 300 volts * 19/200 = 28.5 volts.

But, you have a 25 volt Zener clamping that to probably around 26 volts and that seems likely to be a problem. Transformer action still holds in a flyback design. If the secondary peak voltage is 300 volts, that will be see as a flyback on the primary divided by the turns ratio reversed.

I'm not saying that's your one and only problem but it strikes me as being significant to your issues. If you have a 10:1 step up transformer and your output voltage is nominally 300 volts then, your Zener flyback catcher needs to be in excess of 30 volts (simple math).