Voltage Control On Boost Converter Playing Up / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Voltage Control On Boost Converter Playing Up

previous 1 2 3 next

Move Thread

LAN_403

HV Enthusiast

Sun Jun 04 2006, 01:10PM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

yes

evilgecko

Mon Jun 05 2006, 12:15AM

Registered Member #288 Joined: Mon Mar 06 2006, 07:52AM
Location: Palmerston North
Posts: 32

I added a 1M variable resistor between non-inverting input an output. It seems to fix chattering problem but the charger still fires about 10 times per second which is kinda annoying, and only when I set the variable resistor to about 500K or less.

I thought you add two extra resistors to include hysteresis? As in the diagram the 1M resistor is R2 but I havn't added R1.

What is Vol and Voh in those equations?

Anyways the power of the charger is 8.6W @ 12V. Think I could improve it by making the inductor bigger.

HV Enthusiast

Mon Jun 05 2006, 12:37AM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

You should look into increase the resistance of your voltage divider. It may be discharging the capacitors too fast so that the charging circuit kicks in often.

evilgecko

Mon Jun 05 2006, 01:32AM

Registered Member #288 Joined: Mon Mar 06 2006, 07:52AM
Location: Palmerston North
Posts: 32

The capacitor drops about 1V every few seconds. I trying to get a range of about 5V before it turns on.

evilgecko

Fri Aug 11 2006, 10:55AM

Registered Member #288 Joined: Mon Mar 06 2006, 07:52AM
Location: Palmerston North
Posts: 32

The boost converter is finally built...ok so its been 2 months! One problem though...the mosfet I use for switching keeps heating up.

I'm using IRFP460 which have an on resistance of 0.27Ohms. Is this so much and causing tjhe heat? The trigger signal output from the 555 looks fine with a nice sharp raising edge and falling edge. There is a little ringing after the falling edge at 1500Hz.

At the moment after charging a 1800uF capacitor to 450V it gets very hot.

robert

Fri Aug 11 2006, 12:33PM

Registered Member #188 Joined: Thu Feb 16 2006, 05:18PM
Location:
Posts: 67

The FET will heat a lot but thats normal with this configuration.
It has lousy efficiency and needs overly huge parts. (thats why no one would really consider such high step-up ratio with just a inductor).

Generally, on step-up ratios much greater then 3 or 4 transformers or if no isolation is needed tapped inductors are used.
These converters need a properly designed snubber network across the switching device (thats why most here dont do this, because they dont know how to calculate the part values porperly) but reduce the losses by a great amount requiring much lower voltage switches.

Lets say we go from 10v to 300v.
So we pick a 1:15 transformer, and we have a primary side peak voltage of 20v, which translates to a 30v device, avialible for low prices with miliohm ON-resistances.
Just needs a r-c-d snubber to dissipate the leakage spike that would otherwise damage the switch.

Also, why not use a "standard" current mode controller like the UC384x (though these expect 1.2v drop across the current shunt (designed for off-line apps, with 350v input where 1.2v is insignificant), for low input voltages this requires some offsetting here or amplification to get the losses down)?
These allow clean regulation, current limit and fixed frequency low noise operation with small inductive components.

rupidust

Sun Aug 13 2006, 05:56PM

Banned
Registered Member #110 Joined: Fri Feb 10 2006, 12:23AM
Location: Banned City
Posts: 85

Try a replacement switch. You should be able to charge 180 joules to 450 without excessive heating on the first few consecutive charges. Nothing wrong with using a booster for 12 to 450v amplification. I do it all the time. Try an IGBT or a lower on-resistance IRF part. Actually, just start with an IGBT. Once the regulation kicks in and continues for a minute, the switch should have cooled off. If its still hot, well if its cool during charging and gradually heats during regulation, the issue may be spikes as Robert stated earlier. This would require snubbing, but if so, then its better to get the circuit to work first without snubbing. What you are doing now, Volts, Farads, and Joules wise, is not that critical and should not need snubbing given the charger is working properly. A snubber will be a band-aid, not a cure.

Like I always say, the Inductor much match the frequency. A great booster may on one hand not excessivly heat an IGBT when performing as a 100 watt charger. One the other hand, same said great booster will cook same IGBT in a few seconds if the Induction was to be replaced with another non frequency/power matching Inductor.

What is the charging current?
Is the power output (Joules/charge time) at least 75% of power input (Vbattery * Mosfet current) ? If this is not met, this need to be fixed first.

evilgecko

Wed Aug 16 2006, 05:31AM

Registered Member #288 Joined: Mon Mar 06 2006, 07:52AM
Location: Palmerston North
Posts: 32

How do you measure power input? As I understand DMM's don't work accuratly when measure pulsating charge.

Wilson

Wed Aug 16 2006, 05:57AM

Registered Member #78 Joined: Thu Feb 09 2006, 11:27AM
Location: Sydney, Australia
Posts: 133

I don't think any meters measure pulsating charge properly. I used an analogue ammeter to calculate the power draw from my boost converter, which was charging some caps. And i got some 50W discrepancy between input and output power.....but nothing was heating up....

Steve Conner

Wed Aug 16 2006, 10:03AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Analogue moving coil meters measure average current, which is correct if you are interested in the flow of charge.
Analogue moving iron meters read roughly RMS at low frequencies.
DMMs are supposed to read average, but a highly pulsating waveform could confuse them.

I played with these capacitor charging supplies a while back. I managed to make a good one by hooking the ZVS flyback driver circuit up to the ferrite transformer from an ATX power supply. (using the centre tapped winding that used to supply 5v.) I then rectified the HV winding using a voltage doubler with small caps to limit the current, and hooked this rectified output onto the main cap bank. It would draw about 10A from a 12V supply and charge a 1000uF cap to 400V in a second or two.

previous 1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob