Need to heat an IC...will my solution work?

Status
Not open for further replies.
Got it. Thanks.

Ron H said:
How weird. Did you notice in your post, where you quoted me, it said 160 ohms? That was the value I had given when I thought you were using 9 volts. Maybe I edited it, but there is no indication on my post that I did. Oh well.
Actually, I did notice that. I just figured that while I was working on the post, you had edited your post. I find that some forum software doesn't always add the "Edited..." until some event happens. Either a certain amount of time passes, or the member logs off, or another member reads the post...I don't really know. I just know that if you edit your post quickly after posting it, you don't get the "Edited..." line.

Anyway, I'll start gathering the parts I need for this. Once again, I appreciate all the help.
 
 
Nigel Goodwin said:
it's essential to have some kind of current limiting in the circuit, a simple resistor will do this, and become the heating element itself.
Okay, I understand now. I've already added the resistor into the circuit to be used as the heating element. Now I just have to find those resistors someplace!

Thanks!
 
This is intended to be ran inside a pc is it not? Why are you using a 3.3V regulator and a battery? Just use the 5V power from the PC. (There's nothing in the datasheet that says 5V reduces accuracy)...

Just a note, if you think this heating setup is going to keep your ic at 85°C you're in for a little surprise :roll:
 
Exo said:
This is intended to be ran inside a pc is it not? Why are you using a 3.3V regulator and a battery? Just use the 5V power from the PC. (There's nothing in the datasheet that says 5V reduces accuracy)...
This board will sit next to a PC motherboard that has been specifically setup for testing water-cooling solutions. It will not be permanently attached.

I’m using a separate battery because the first sensor circuit I made exhibited strange and irresolvable behavior that appeared to be related to grounding, so I wanted an isolated power system for the sensor.

I’m using 3.3V because that’s what’s specified as the conditions required for +/-0.8C accuracy. There is a parameter called “Supply Sensitivity of Temperature” that’s given as +/- 0.2 C/V. Seeing that voltage can affect temperature, I figured I’ll just stick with the specs.
Exo said:
Just a note, if you think this heating setup is going to keep your ic at 85°C you're in for a little surprise :roll:
Why is that?
 
I was looking at the circuit and I was wondering...

Since I'm using a battery for power, shouldn't I be able to remove both 0.1uF caps and the 200 Ohm resistor without ill effects? Those are for noise reduction, which I'd think I shouldn't need in this particular setup. The output from the 1658 seems clean enough to not cause any problems, and there's still the 10uF cap across the output supply.

SHDN is just a logic level. Since I have no shutdown circuitry, I can just connect SHDN directly to Vin, right?

Thanks.
 

The 0.1uF's provide a low impedance path across the supply, and the 200 ohm forms part of a low-pass filter - I wouldn't suggest leaving them out. A battery supply needs decoupling capacitors just as much as a mains supply.

SHDN is just a logic level. Since I have no shutdown circuitry, I can just connect SHDN directly to Vin, right?

Should be fine!.
 
Follow-up...

The circuit hasn’t worked yet.

First, the 9V batteries were dieing quickly, so I started using a large 12V battery instead. I used a 10 Ohm Vishay metal-film resistor as the heater because it has a heatsink mount and proper power rating. However the temperature sensor only reached 67C with it. Also, the 1.3k Ohm resistor was way too large to get enough current to flow and was eventually reduced to 270 ohms.

The last thing I tried was, essentially, my original idea except using the PN2222A instead of a JFET. That worked in getting the IC up to 85C (was able to go over 90C.) However, something was preventing the OVERT signal from working properly. Sometimes the circuit would turn off the transistor, sometimes it wouldn’t. So the temp would go up, down, up, and then either just keep going up, or go down and never come up again. The behavior was random. I think it has something to do with the amount of current I was forcing through the OVERT pin.

I think the next thing to try is to use the OVERT signal to turn on a PN2222A that is controlling a second PN2222A to heat the sensor. I should be able to greatly reduce the current flow through OVERT that way.

Another option is to use the unregulated input voltage for the transistor-resistor heating circuit. If the resistor was able to heat the IC to 67C on 3.3V, it will definitely go to 85C on 12V.
 
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…