[EE] Use triac to switch tens of kHz ?
Peter P.
can a triac be used to switch a higher frequency signal ? By this I mean
asynchronous, on/off switching, probably with DC drive on the gate. Basically is
there anything that prevents a triac from turning off properly when its A1-A2
see maybe 50kHz and 50Vpp ? Power electronics texts yield no clues and triac
manufacturers likely don't test for this at all. I am aware of horizontal
deflection circuits in TV implemented with thyristors (of somewhat special
make), but this is different, I will switch the triacs at a much lower frequency
than the 'power' input.
thanks for any pointers,
Peter P.
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
Vasile Surducan
could turn on the triac without any gate command. So the load
characteristics it's important (it's resistive or inductive?). On old
TV the 16KHz horizontal deflection use two thyristors just to prevent
this problem but the voltage was huge compared with your application.
The question is why use triac at 50V and 50KHz switching (assuming
you're not switching 100 A) ? The reason of using a triac is
automatic turn off at lowest A1-A2 current than holding current and
bidirectional current conducting.
Using two MOS FET's will do the probably same like a triac without
those problems.
Peter P.
Vasile Surducan <piclist9 <at> gmail.com> writes:
> AFIK, there are some issues on fast triac switching like dv/dt which
> could turn on the triac without any gate command. So the load
> characteristics it's important (it's resistive or inductive?).
I thought about dV/dt but my dV/dt is not so terribly high. The input is a
reasonably almost-sine wave from a dc/dc converter through a simple filter.
> The question is why use triac at 50V and 50KHz switching (assuming
> you're not switching 100 A) ? The reason of using a triac is
> automatic turn off at lowest A1-A2 current than holding current and
> bidirectional current conducting.
> Using two MOS FET's will do the probably same like a triac without
> those problems.
I need to switch about 8 Aeff. So the peaks could be 100A at least temporarily.
Using 2 inexpensive mosfets 'back to back' is possible but I don't like what I
need to do to drive the gates (the gates float on the 50kHz - it is a high side
switch). That is: I know precisely what I need to do to drive the gates and I
have done it before, but I would like something less sensitive (than mosfets) as
a switch there.
Peter P.
> I thought about dV/dt but my dV/dt is not so terribly high. The input is a
> reasonably almost-sine wave from a dc/dc converter through a simple filter.
Thinking further on this, there are dual gate thyristors which allow the dV/dt
to be controlled a little better, like the low power BRY39. Are there such
things for higher power, in the triac class ?
fred jones
I am building a project that has some high gain, low noise amplifiers and
will need to be in a shielded housing. My first choice was to buy a metal
enclosure but I could not find one that fit other requirements. I ended up
building it around a Hammond mfg 1598J plasic enclosure. Can anyone suggest
a way to shield the inside of the box? Is there some type of spray that I
could spray inside to do this? The only requirement of the shielding
material is that it has to be non-ferrous. I will be discarding the plastic
end panels and having some aluminum ones machined. Thanks so much for any
suggestions.
FJ
_________________________________________________________________
Download Messenger. Join the i’m Initiative. Help make a difference today.
http://im.live.com/messenger/im/home/?source=TAGHM_APR07
Kenneth Lumia
Could you specify the amount of shielding needed? Although
coated cases can sometimes help, they do have a tendency to leak
around the joints. You may want to look at doing a shield can on
the most sensitive parts of your circuit, and then decouple the
rest of the circuitry as appropriate. We really need more
information, specifically what signals are going in/out of the
box, and how much shielding is required.
Ken
--------------------------------------------------------------------------------
Jonathan Hallameyer
can be used for shielding, but it might be impractical to use that if this
is being produced in quantity.
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>
>
--
Jonathan Hallameyer
Russell McMahon
> amplifiers and
> will need to be in a shielded housing. My first choice was to buy a
> metal
> enclosure but I could not find one that fit other requirements. I
> ended up
> a way to shield the inside of the box? Is there some type of spray
> that I
> could spray inside to do this? The only requirement of the
> shielding
> material is that it has to be non-ferrous. I will be discarding the
> plastic
> end panels and having some aluminum ones machined. Thanks so much
> for any
> suggestions.
You can (or could and almost certainly still can) buy Nickel loaded
spray specifically sold for shielding purposes. Non ferrous shields
(whether sprayed or solid material) will provide electrostatic
shielding but for electromagnetic shielding you need the real deal. I
haven't tried it for shielding but you can or could but zinc loaded
paint for "cold galvanising" and if the weight of the tin is anything
to go buy the metal content is reasonably high. You'll note that some
Hammond plastic enclosures already have a shielding spray on the inner
surfaces.
For especially problematic equipment that demands electromagnetic
shielding you can build shields out of sheet steel which may be
masquerading as tin-plate wedding cake slice boxes (probably nowadays
as rare as Dodo teeth), stove element protectors (pretty round things
with pictures on), suitably sized "tin cans" or just good old
galvanised sheet steel (which is generally thicker and so harder to
cut). Most of these do a good job of cutting unwary fingers when cut
and bent to suit. Tin cans and tinplate in general has the advantage
of being reasonably possible to solder without a blow torch. (A Weller
etc will almost do). Galvanised steel is usually less easy.
For really demanding applications mu-metal is the super miracle
material, can be bought currently and has the super miracle price to
match. Some old surplus valve equipment has mumetal valve shields but
this is also probably getting down to the Dodo teeth level.
Russell
Alan B. Pearce
Old fashioned cake or biscuit tins may also suit this role. The main part of
the tin may need trimming down to a suitable depth, but for a one-off could
be ideal. May also be able to get rectangular versions that might suit some
form factors better.
Another "ready made" one to remember is the zinc alloy boxes made in the UK.
These come in similar sizes to many of the smaller plastic boxes, and are
very easily machined.
PAUL James
All,
Another idea that may have already been suggested is to cut pieces of
blank PC board to size, and solder them together to fit inside the
plastic box. Again, this is okay for onsey twosey type things, but for
production quantities, it would be difficult at best, and not really
practical.
Regards,
Jim
Scott Touchton
operating between 7 and 15Khz. Peak current is around 1 amp, voltage is
around 150V due to the Q of the network. Triac is turned on and off by a
PIC I/O controlling the gate.
If interested I can look up and let you know the exact part I am using.
Scott
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
can a triac be used to switch a higher frequency signal ? By this I mean
asynchronous, on/off switching, probably with DC drive on the gate.
Basically is
there anything that prevents a triac from turning off properly when its
A1-A2
see maybe 50kHz and 50Vpp ?
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.5.463 / Virus Database: 269.5.9/773 - Release Date: 4/22/2007
8:18 PM
Peter P.
> I use Triacs to select tuning capacitors in a series resonant circuit
> operating between 7 and 15Khz. Peak current is around 1 amp, voltage is
> around 150V due to the Q of the network. Triac is turned on and off by a
> PIC I/O controlling the gate.
>
> If interested I can look up and let you know the exact part I am using.
Thanks for answering. That seems to confirm that it's possible. There is no need
to specify the parts you use, just whether they are ordinary or special. I
looked at some datasheets and the dV/dt I will be seeing is well within spec
(10x and over) of what these parts can do. Any special measures ? Holding the
gate with a low value resistor ?
thanks,
Scott Touchton
held to ground via a .1uF cap, and fed from a PIC I/O pin via a 330 ohm
resistor.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Any special measures ? Holding the
gate with a low value resistor ?
thanks,
Peter P.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Peter P.
> I use a sensitive gate triac that fires in all four quadrants. The gate is
> held to ground via a .1uF cap, and fed from a PIC I/O pin via a 330 ohm
> resistor.
Ok, thanks. I hadn't thought about a cap there but now it makes sense. Normally
one holds the gate to K with 50 ohms and it works. I'll try a capacitor too and
see what happens.
Scott Touchton
I went out on limb a little with the circuit, but could easily watch the
conduction and shutoff of each triac in the tank. Needless to say, about
8000 units later and 2 years of operation in the field all seems well. I am
not a thyristor guru by any stretch of the imagination, so struggled a
little getting this design off the ground. Probably had a good dose of luck
tossed in too.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Ok, thanks. I hadn't thought about a cap there but now it makes sense.
Normally
one holds the gate to K with 50 ohms and it works. I'll try a capacitor too
and
see what happens.
thanks,
Peter P.
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
No virus found in this incoming message.
Russell McMahon
> of
> blank PC board to size, and solder them together to fit inside the
> plastic box. Again, this is okay for onsey twosey type things, but
> for
> production quantities, it would be difficult at best, and not really
> practical.
Works well for electrostatic shielding. No good if you want
electromagnetic shielding.
You can also buy brass foil (shim stock ?) and probably copper foil as
well.
Russell
John La Rooy
>
>
> For really demanding applications mu-metal is the super miracle
> material, can be bought currently and has the super miracle price to
> match. Some old surplus valve equipment has mumetal valve shields but
> this is also probably getting down to the Dodo teeth level.
You can find little slabs of mumetal in hard drives, Older ones are usually
easier to open up. The magnets are also very strong glued to the mumetal.
John