Overclock.net › Forums › Cooling › Specialized Cooling › Peltiers / TEC › TEC temperature controller
New Posts  All Forums:Forum Nav:

TEC temperature controller - Page 3

post #21 of 72
yip
post #22 of 72
Quote:
Originally Posted by Nighthawk82 View Post
The circuit you soldered with the mosfets, isn't that actually some sort of solid state relay?

"A solid-state relay may include a driver circuit and a field-effect transistor (FET) providing the output signal and having a control terminal or gate connected to the driver circuit. The driver circuit receives the control signal and operates or drives the output FET based upon the control signal."

The only difference I can find at first sight:
"A typical solid state relay consists of a light-emitting diode (LED) optically coupled across an electrically isolating gap to a photovoltaic array. The photodiode array is electrically connected to an output device such as a field effect transistor (FET)."

So you control the mosfets directly with the control signal, while in a solid state relay the control voltage is "optically coupled" to the mosfets so that there's isolation between the control circuit and the switched circuit.

So I'm just wondering, wouldn't an SSR be just as good as a solution, or even better because of the isolation of the control and switched circuit? Since the SSR is also consists of mosfets. And an SSR switches silently, so there shouldn't be any noise.
the mosfet obviously isolates the two as well.the relay may very well be just fine to use the you just need to check it's speed

if it does work it Could make life some what simpler. then you may not be affected by the cap issue or it may be resolved with some resisters . It's worth looking into. Of course it may not fix anything

And it's been a pleasure to help so far
post #23 of 72
Thread Starter 
While we're at it...
In your post you sketched a problem with a voltage drop with 2 circuits.





You mentioned that the second one resolved all your problems, may I ask how? Is there such a big difference in the placement of the TEC in your circuits above?
post #24 of 72
20 milliseconds to turn on and off is pretty slow for a SSR
Where are the Mosfet is 140 ns
i couldn't see the SSR working. anyway i see no reason to use the SSR anyway

http://newzealand.rs-online.com/web/...duct&R=5352968
http://docs-asia.electrocomponents.c...6b807910db.pdf
Time conversions
http://www.simetric.co.uk/si_time.htm
post #25 of 72
Quote:
Originally Posted by Nighthawk82 View Post
While we're at it...
In your post you sketched a problem with a voltage drop with 2 circuits.





You mentioned that the second one resolved all your problems, may I ask how? Is there such a big difference in the placement of the TEC in your circuits above?
the load is one the wrong side . which stuffs every thing up.

the TEC needs to be on the Drain side not the Source side not surprisingly enough
Edited by Ultrasonic2 (muffy) - 3/17/09 at 6:57pm
post #26 of 72
Thread Starter 
Quote:
Originally Posted by Ultrasonic2 (muffy) View Post
20 milliseconds to turn on and off is pretty slow for a SSR
Where are the Mosfet is 140 ns
i couldn't see the SSR working. anyway i see no reason to use the SSR anyway

http://newzealand.rs-online.com/web/...duct&R=5352968
http://docs-asia.electrocomponents.c...6b807910db.pdf
Time conversions
http://www.simetric.co.uk/si_time.htm
Indeed, now that you mention it. Strange that an SSR is so slow in comparison, although it's also built with mosfets.
Do you have any idea how fast the switching frequency is with your t-balancer? I don't really know if the frequency should be in that fast range, it seems really fast. Going in the range of like 10ms would yeald the same results I would say, since eventually it's the average current that is applied. Also, letting it switch that fast, doesn't it shorten the lifespan of the mosfets drastically too? I don't know an exact figure for the switching lifespan while I'm writing this, but if yours are switching that fast then wow
post #27 of 72
im sure it would take AGES to stuff a mosfet from switching it since that's what it's made to do. I can find out the switching frequency of the T-balancer since it's adjustable when i get home.

From memory it's max it just below the min recommenced for TEC's

Why are you so obsessed with using a SSR anyway ? there like 5 times the price of a good mosfet anyway ?
post #28 of 72
t balancer does 828hz

with the relays 20ms for on and off that would give it the fastest possible time of 50hz but at 50 it would spend no time on as it would be turning off again i guess.

where as the mosfet could obtain 7,142,857hz
Edited by Ultrasonic2 (muffy) - 3/17/09 at 9:05pm
post #29 of 72
Thread Starter 
Quote:
Originally Posted by Ultrasonic2 (muffy) View Post
Why are you so obsessed with using a SSR anyway ? there like 5 times the price of a good mosfet anyway ?
Hehe, not obsessed, you already convinced me in the beginning, I just like to know all options and why it would or wouldn't be a good choice, you never know that kind of knowledge may come in handy later and prevent you from making stupid decisions
post #30 of 72
Thread Starter 
Here is some interesting stuff regarding TEC temperature control using PID, what I was referring to before, it also mentions something about TEC response time. But skip the mathematical stuff
Thermoelectric Cooler (TEC) Control

PWM Temperature Controller for Thermoelectric Modules Keeps Components within 0.1°C

Ok it's for TEC used for diode cooling but it's still interesting
They actually use a H-bridge to be able to cool and heat, so inverting the TECs polarity for better temperature control.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Peltiers / TEC
Overclock.net › Forums › Cooling › Specialized Cooling › Peltiers / TEC › TEC temperature controller