Boost converter gone wrong

[ronsimpson]

Don't the 555 need a high voltage spike protection came from the inductor?

no
What spike? I don't see a way for the high voltage to get back to the 555.
There needs to be a power supply cap on the 555. Some of the problems might come from the supply voltage changing because there is no cap.

 

[eyeofjake]

So, I broke out the bread board and assembled the circuit. Added a cap at the source and a 50 ohm resistor to the gate. At first it didn't work so I tried different inductors. I hooked up a huge transformer on the primary and it worked.....for a few times and I'm back to square one.

 

[eyeofjake]

OK, so I made some progress:

Realized that I had some poor connections at the 555 killing the pulse to the MOSFET. The timer is now working fine.

Once I got it all hooked up I used an IRF740 MOSFET as the switch. I made an adjustment to the schematic and added a variable resistor at both R1 and R2 to play with the duty cycle. I had to use a pretty heavy duty toroid inductor I made to get any decent charge build up. This baffles me as I have seen other circuits use smaller inductors with better efficiency. One circuit design I found used a small 200 uH coil to achieve 300 volts. The timer is remains cool but the MOSFET gets hot. Also, if the coil is big enough, I get a huge voltage spike according to my voltmeter as it will read something like 1700 volts but when I kill the circuit the cap array is only at 120.

So, to remedy the hot MOSFET, I upgraded to an IRFP450 but it does switch on. I thought it was bad then tried another one but nada. The circuit works with the IRF740 though I can't find an adequate duty cycle that charges to 300 volts in a stable manner. I have compared my circuit to other designs and I can't see what is wrong. According to other circuits I have found online, this should be working yet isn't.

I'm at a loss here as I calculated that I need a 95% duty cycle for a small inductor and from there I got the resistor values. Yet, they didn't work and especially with the recommended MOSFET (IRFP740).

 

[Diver300]

I think that you would do better with a dedicated boost converter controller IC.

With your circuit it's difficult to work out when you have saturated the inductor. If you use something like this:-
https://datasheets.maximintegrated.com/en/ds/MAX770-MAX773.pdf
there is current measurement in for the MOSFET current, so that the peak inductor current can be limited.

 

[Pommie]

Can you correct your diagram as I believe C1=0.001pf is incorrect as you can't buy capacitors that small. What frequency do you think you're working at?

Mike.

 

[Pommie]

As an after thought, why use such a high and dangerous voltage? A bank of three 20,000μF capacitors charged to 24V will still give you 17 Joules. Plus, you could buy a boost converter off of ebay for about $1 that will do everything you need. Something like **broken link removed** with a current limiting resistor (12Ω) will achieve what you require and be a lot safer. You can adjust the power level with the variable voltage and it will recharge in around 1 second.

Mike.

 

[eyeofjake]

So I can't use 24 volts as this needs to be portable. Carrying around a bank of super capacitors is a nightmare. The issue with buying premade ones is that they would have to be retrofitted into a housing. I rather build it custom to a size that way I can have total control of the system and also know what to fix if it is faulty.

I'm aiming for 20kHz with a duty cycle of approximately 95% according to the math I did. The minimum inductor value is about 800 uH so my 2 mH inductor should be sufficient.

I'm still having a tough time figuring out why one MOSFET works but the other doesn't. Both have a minimum gate voltage of 2 volts according to the data sheet yet the power mosfet doesn't wont to switch on. The IRF740 does so without a problem.

 

[eyeofjake]

diver that chipset looks rather interesting. Basically an all in one set up. I would have to tweak it to get higher voltage but should be doable.

 

[dr pepper]

I just tried to run a calc on the 555's timing components, you list .001pF for the timing cap, this cant be right, even if its 0.001uF that woudl still produce an o/p freq of 500khz, which is way too high.
2mH sounds very high for a smps of this kind.
Whats your intended frequency of operation?
2mH will only pull a very low current at typical smps operating frequencies.
Heres a good site to gain some knowledge on switchers, and theres a high voltage regulated supply example, changing R4 on fig3 ought to do it, a 1000v fet might also be a good idea.
https://www.dos4ever.com/flyback/flyback.html

 

[davenn]

As an after thought, why use such a high and dangerous voltage? A bank of three 20,000μF capacitors charged to 24V will still give you 17 Joules.

So I can't use 24 volts as this needs to be portable. Carrying around a bank of super capacitors is a nightmare.

I really don't think you understood Pommie 's comment ... no one is talking about supercaps

and you still haven't commented on the incorrect capacitor value


Dave