Mail Archives: geda-user/2018/02/16/11:10:00
Hi Marcel,
Op 16-02-18 om 16:10 schreef mhx AT iae DOT nl:
> On 2018-02-16 12:38, Richard Rasker (rasker AT linetec DOT nl) [via
> geda-user AT delorie DOT com] wrote:
>> Hello Marcel,
>>
>> Op 16-02-18 om 06:32 schreef mhx AT iae DOT nl:
>>> On 2018-02-15 23:02, Richard Rasker (rasker AT linetec DOT nl) [via
>>> geda-user AT delorie DOT com] wrote:
> [..]
>> For all clarity, I attached the schematic, also because there's
>> something else I don't understand: the top MOSFET (Q1) doesn't get the
>> burst frequency at all, just the 1.5 ms LF pulses. The bottom one (Q2)
>> is supplied with the burst signal as expected.
>
> That must indeed be wrong. But there is something else, the top and
> bottom driver
> want 180 degrees out of phase signals. An inverter seems to be
> missing, or
> the two exors are wired differently than you think?
> [https://wwwee.ee.bgu.ac.il/~pemic/publications/jour3.pdf]
I just received an answer from the manufacturer, and there was indeed an
error in their documentation.
U1 pin 1 should be connected to the 41 kHz signal, not to the LF pulse.
> No, this looks like a push-pull allright. Are those two 1uF caps not
> connected
> to the center tap? The electrolytic 4700uF might not like the very
> large AC ripple.
The 1uF caps are definitely connected between drain and source, and from
the PCB view, it would appear that they're pretty standard 0805 ceramic
types, so no plastic film or other high-quality types. And even the
1.5kV capacitor on the secondary is still a 1210 SMD type (although I
probably go with a through-hole film type there).
>> BTW, the circuit isn't really designed for resonance, as the 41 kHz
>> given is in reality a selectable frequency between 3 kHz and 300 kHz,
>> with 41 kHz being just one of the settings.
>
> That is certainly going to blow when built like this, then. To make it
> work
> between 3 and 300 kHz should be impossible. With that specification
> you'll want a
> linear amplifier. With a dutycycle of 1.5ms/10s = 1:6666, dissipation is
> not the issue, when the SOAR of the devices is respected.
It actually appears to work -- here's a public document describing the
thing: http://www.airmartechnology.com/uploads/InstallGuide/17-143.pdf
If they indeed manage to pull it off with this somewhat unusual design,
then kudos to them -- and appearances can be deceiving. I've seen
'heavy' designs buckle at the knees where lighter, elegant stuff sailed
through just fine.
I'll use it as a starting point and reference, and see how it actually
performs; I'm almost tempted to order one and see for myself, although
building it from the schematic isn't much work either; the most critical
part is the transformer, and I have good specifications for that one.
I'll also try and see if I can simulate this in LTSpice, although then
I'd first have to learn how to model this transformer properly. One
thing I can say: It's already quite an education.
Anyway, thanks again for your input and support!
Met vriendelijke groeten,
Richard Rasker
--
Linetec
Vleerkottelanden 14
7542 MJ Enschede
053-4350834
http://www.linetec.nl/
- Raw text -