TL494 Modulation Problems / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

TL494 Modulation Problems

1 2 next

Move Thread

LAN_403

Richard

Sun Jan 13 2008, 09:57PM

Registered Member #1211 Joined: Fri Jan 04 2008, 11:20AM
Location: England - UK
Posts: 19

Hello everyone.

I'm trying to build a decent quality plasma 'arc' tweeter...

I've knocked up a simple Flyback/LOPT, driven by a 555 / IRF740, running at approx 120Khz.

The arc is fairly broad, about 1mm - quiet & steady at 30mm, with a faint hint of whitening.
The audio signal is fed to pin 5 (pulse position modulation) after been first amplified / High-pass filtered by an op-amp. The resulting audio quality is OK, but seems 'flat' & suffers with some distortion..

Having read some very interesting HV 'audio' articles, about using the TL494/UC3525 for PWM, I thought I'd have a bash at using a 494 & modulating it (deadtime), to get a side-by side comparison of the two.

With the duty cycle set at approx 85% @ 120Khz (as with the 555), I get the same arc - no problems.
Connected an audio source to pin 4 - nothing. Audio is present on the input, but no flame modulation.
Backing off the duty cycle greatly, results in some very nasty distorted audio, with a weak 'plasma' flame.

It may be of some relevance, maybe not, but I am using a totem-pole / push-pull pair, driving a single power MOSFET, which is snubbered, all fed from pins 9 & 10 (joined outputs) of the TL494.

Many thanks for any ideas :)

Richard (England)

Mates

Sun Jan 13 2008, 11:09PM

Registered Member #1025 Joined: Sun Sept 23 2007, 07:53PM
Location: Czech Rep.
Posts: 566

timetec wrote ...

The resulting audio quality is OK, but seems 'flat' & suffers with some distortion..

In my opinion, plasma tweeter made of fly-back will always sound like shit. I'm surprised you can even run a fly-back at 120KHz...You can check this scheme for audiomodulation of tl494 - it may help...

Once it plays I'd really like to hear the result...

Cheers Mates

Richard

Mon Jan 14 2008, 12:20PM

Registered Member #1211 Joined: Fri Jan 04 2008, 11:20AM
Location: England - UK
Posts: 19

Thanks for the reply.

I've got the very same circuit in front of me and it was one of the variations I used to build the TL494 PWM driver, others been the one that Jan Martis (jmartis) devoloped, along with the Michi and VCore designs.

Something's very wrong - the only thing different, is I'm using a one FET driver, not a bridge.
If I crank up the audio input to pin 4, with a steady arc, I can just hear some very distored, severely clipping audio, in the background. The other thought, is that most of the designs use a dedicated Gate driver IC for the power FET... however, I had not heating / slow switching problems, using a totem-pole driver in conjunction with the 555 - nice clean square waves all round & fairly good audio ?

Kizmo

Mon Jan 14 2008, 01:42PM

Registered Member #599 Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 624

I made one few weeks ago

- TL494
- IRFP460 mosfet
- 9A UCC gate driver
- ATX PSU for TL494 + gate driver UCC
- Transformer and some capacitors to make ~30V for mosfet to switch

It is just TL494 square wave generator @ 20kHz which drives 9A UCC gate driver IC which drives gate of IRFP460. Audio is fed like this:

I think audio modulated flyback driver cant be much more simpler than this. And reliable too, hours of playing and no failures.

Video http://www.youtube.com/watch?v=1Blrhl0pRIs

Richard

Mon Jan 14 2008, 03:07PM

Registered Member #1211 Joined: Fri Jan 04 2008, 11:20AM
Location: England - UK
Posts: 19

Hi Kizmo

I must have watched your video a dozen times - excellent work !

Do you think a dedicated gate driver(s) is essential for this circuit operation ?
If so, I'll have to get some bought - ditch the totem-pole driver.

Also see you are using a UCC driver, not a a standard TC4420/29 - any difference ?

I've kept well away from bridge designs so far, in so much as I can't source the required HV 2-10uf capacitors, needed to drive the primary winding.
The one attempt at a half-bridge I did cobble together, used some low value devices - no drive whatsoever and hence no output :(

Re. the picture link - thanks very much , I'm using the same technique (think it was on the scopeboy.com site), as a template for the deadtime modulation method.

However, not all doom & gloom - I've had much greater success FM modulating the TL494 !
Connecting a high value resistor (22K in this case) from pin 7 (Rt) to the audio source, through a 0.22uf cap.
Fairly quiet, but really good sound quality - you might want to give it a go.

I'll post some pictures & possibly a video, as soon as I get chance.

Thanks again.

Dr. Dark Current

Mon Jan 14 2008, 04:09PM

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

Hi Richard, I think your problem is that you use only one FET- with the bridge, the output is much more "linear". Also to get decent audio levels, there must be quite a bit of power in the arc, I guess mine was ~60W.
The caps are not so expensive, actually any film or foil cap will do.

If you built my circuit and tune it well, it should give a relatively clean output at moderate volume. The key is to set the frequency and duty-cycle bias pots well.

If the sound is too low, an additional preamplifier can be used. Alternatively, primary winding turns can be decreased but just until the core starts saturating, you'll notice this for sure as the audio gets very squeaky.

Oh I forgot- be sure to use the old-type flyback with AC output. The new DC one might work too but the audio quality can be decreased.

Good luck, J.M.

uzzors2k

Mon Jan 14 2008, 08:19PM

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

Gate driver chips are absolute over-kill at such low frequencies, especially when not using a GDT. You should be able to make it work with what you have available right now. The way you connected the TL494 outputs sounds fishy to me, can you post a schematic? At any rate, for single ended use, pin 13 must be pulled low. Also I would ground pins 9 and 10 so the TL494 only has to source current. Here is an example of how to do it.

1200341793 95 FT37378 Superiour Control Flyback Driver

Then just hook up the audio input as shown on Steve's schematic. As Jan said the audio output might need amplifying for the duty cycle to be changed noticeably.

Kizmo

Mon Jan 14 2008, 08:38PM

Registered Member #599 Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 624

I used that gate driver just because i have lots of them lying around

...

Tue Jan 15 2008, 12:34AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

@kizmo, you should not be pulling arks to a drill bit like that... If you had one side of the flyback grounded (or the flyback managed to ark to something grounded) your body would be floating at several thousand volts, and probably flowing considerably RF current.

@Richard, I think that you would have a lot better luck using a halfbridge, as with a single fet the majority of the output is being created by inductive kickback and won't be affected by the duty cycle. A halfbridge should fix that.

Richard

Tue Jan 15 2008, 08:30PM

Registered Member #1211 Joined: Fri Jan 04 2008, 11:20AM
Location: England - UK
Posts: 19

Thanks to everyone for your advice - Have ordered a selection of suitable caps / IRF540's & 740's for the half-bridge design. With a bit of luck, I shall have a working driver (with audio) by the weekend :)

Some time now, to have a go at building Uzzor's TL494 version - cheers Uzzors.
It's similar, except it looks like the O/P transistors are connected 'inverted' - others I've seen have the output driven from the tied emitters - interesting stuff.

1 2 next

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