Variable voltage flyback driver. / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Variable voltage flyback driver.

1 2 next

Move Thread

LAN_403

Arcstarter

Sun Mar 13 2011, 01:53AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

So, i am looking for a variable voltage flyback driver, probably just 0-30kv. It will be for X-ray experiments, but im not sure what the best way to get this is.

I am thinking current mode, with a typical current mode driver, which, in my case, would be the uc3842. I also have some TL431 shunt regulators, as well as some higher voltage (~80v) shunt regulators.

My question is, how would i get feedback, seeing that it is such high voltage? I have a few ideas...

Lets say i use a schematic similar to Uzzors':

One thing i could do is use the same feedback schematic with the opto and shunt regulator, but with an external 12v source. Then, just use the correct ratio voltage divider for the reference voltage, one of which resistors being a pot. A 220M resistor and a 20K pot gives 0-2.7v, but would the reference pin of the TL431 pull enough current that the output of the resistive divider would be way off? With the typical reference current being 1.8ua, i don't think so, seeing that at 30kv the current through the divider would be 136ua with said resistors anyway.

As for the power section, i would like to use 170v into the flyback so that i do not need a transformer, but there is not a winding on most flybacks for that voltage. Im not sure what to do about that...

If i really have to, i could use a linear variable voltage supply with an op amp and big transistor and just some typical driver.

What do you guys think?

Inducktion

Sun Mar 13 2011, 04:53AM

Registered Member #3637 Joined: Fri Jan 21 2011, 11:07PM
Location: Buffalo, NY
Posts: 1068

Arcstarter wrote ...

So, i am looking for a variable voltage flyback driver, probably just 0-30kv. It will be for X-ray experiments, but im not sure what the best way to get this is.

I am thinking current mode, with a typical current mode driver, which, in my case, would be the uc3842. I also have some TL431 shunt regulators, as well as some higher voltage (~80v) shunt regulators.

My question is, how would i get feedback, seeing that it is such high voltage? I have a few ideas...

Lets say i use a schematic similar to Uzzors':

I don't think the optoisolator would survive 30 kv in the secondary.

And, you could probably use a ZVS driver with just a variac, or a variable power supply. It shouldnt be that hard.

Arcstarter

Sun Mar 13 2011, 05:35AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

The opto would not see anything higher than 12v. The hv would go through a resistor divider, and the highest voltage that it could output is 2.7v.

I do not have a variac, and a variac isn't solid state. I want this to be purely solid state, mostly because that's what i have, but also because it is more efficient and smaller.

Sulaiman

Sun Mar 13 2011, 11:57AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

Most T.V. flybacks operate from a dc supply of around 130V (+/- quite a bit) so 170 Vdc is probably a good match for the original primary.

You don't need to isolate the eht -ve voltage, you can 'earth' the bottom of the secondary and the logic supply, so an opto-isolator isn't required and feed back the output voltage via the resisiove divider directly to the smps ic.

If you do want to isolate the output you will probably find a part of an auxiliary winding on the flyback transforner that is suitable to make an isolated low voltage dc source.

Antonio

Sun Mar 13 2011, 01:59PM

Registered Member #834 Joined: Tue Jun 12 2007, 10:57PM
Location: Brazil
Posts: 644

This power supply is similar to the ones that can be found in most electronic devices that use a switching power supply. I have taken complete power supplies like this from discarded monitors. I am not sure if a HV flyback transformer can be safely used instead of a specifically designed power transformer, specially if a highly variable load is present at the HV side. Note that the primary side is directly connected to the power line, so the optoisolator is necessary. As Suleiman said, the regular solution for sampling the output is to use one of the other windings of the transformer. There is some imprecision on doing this, however.

uzzors2k

Sun Mar 13 2011, 03:02PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

I'm actually working on the exact same thing Matthew, have you seen Jan Martis' current mode flyback driver?

I've made a PCB for it, but it's tailored to the components I had at hand, so I'm not sure how useful it is. Let me know if you want it.

Anyway, by using the power potentiometer you can rather smoothly adjust the output voltage from zero to max, which has been around 40kV for the flybacks I've tested. I haven't taken the time to regulate the output voltage as the load from tube shouldn't fluctuate much. If you do choose to regulate the output, ground the flyback and you can cut out the opto and TL431. Simply feed your voltage divider into VFB. Maybe put a bypass capacitor on the compensation resistor, 10nF or something.

Btw, the UC3842 won't startup from 12V, so get a 18V supply.

Dr. Dark Current

Sun Mar 13 2011, 06:08PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

If you want to regulate the output voltage, you can take current feedback from the bottom of the Focus&Screen divider. The signal will be probably very small so you may need to amplify it before going into the SMPS control IC.

Adam Munich

Sun Mar 13 2011, 08:43PM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

Actually. if you can vary the frequency of a half bridge driver that serves to vary the voltage as well. Lower Hz = Higher volts, less mA. Higher Hz = Less volts, more mA.

Use this, should work. You can also provide feedback to the IC with a big voltage divider.

Arcstarter

Mon Mar 14 2011, 01:38AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Uzzors wrote ...

I'm actually working on the exact same thing Matthew, have you seen Jan Martis' current mode flyback driver?

You know, im not sure why i thought i needed a regulated output. A driver like this would be nice, i already know how to use the IC as it is use in an ATX supply, but i never thought to use pots for PWM. So, that solves one problem, and as Sulaiman stated, i could find an LOPT with a primary voltage to match or at least get close to mains.

But, the tube i am using only pulls around 50-100ua in setups others have made due to the rated dissipation being 2.69 watts. Would i be able to smoothly adjust the output voltage even though the load is low? I guess the main problem would be using a pot that is precise enough to vary the voltage smoothly even if the load is low, but that could just be a second pot in series with the first pot that has low resistance for more accuracy.

Thanks for all the input guys! Now i just need to find a way to get a flyback's pin spacing onto PCB express without a caliper...

James

Mon Mar 14 2011, 02:25AM

Registered Member #3610 Joined: Thu Jan 13 2011, 03:29AM
Location: Seattle, WA
Posts: 506

If you're driving it in flyback mode, you can use another secondary winding to monitor the output. The voltage on any additional secondaries will closely track that on the main output. You can also use a resistor divider to sample the output directly, I've worked with a dental head that uses a 200M resistor to do this.

30kV is a bit low for xray work though. 50kV is about where things start to get interesting,

1 2 next

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