• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Hysteresis

Status
Not open for further replies.
Hi,

I have built this fridge thermostat (220V/12V/Gas absorption type fridge) which works perfectly but have a question or two.

- R6 sets the hysteresis, is it possible to calculate this value? R6 is currently 500k and when the temperature is selected it gives the set temperature +0.7°C, on reaching +0.7°C will drift back down the set point. I would like to decrease the range but experimentation is time consuming and would like to avoid this.

- I have taken my initial prototype board (waste not want not) and have converted it to operate the 12V element, at 175W this is a draw of 14,6A (only need this for limited durations, 4 hours or so). Is there such a thing as "starting current" with a resistive load? If so will an increased amount of switching increase the current consumption?

Regards
Andrew
 

Attachments

ericgibbs

Well-Known Member
Most Helpful Member
Hi,

I have built this fridge thermostat (220V/12V/Gas absorption type fridge) which works perfectly but have a question or two.

- R6 sets the hysteresis, is it possible to calculate this value? R6 is currently 500k and when the temperature is selected it gives the set temperature +0.7°C, on reaching +0.7°C will drift back down the set point. I would like to decrease the range but experimentation is time consuming and would like to avoid this.

- I have taken my initial prototype board (waste not want not) and have converted it to operate the 12V element, at 175W this is a draw of 14,6A (only need this for limited durations, 4 hours or so). Is there such a thing as "starting current" with a resistive load? If so will an increased amount of switching increase the current consumption?

Regards
Andrew
hi Andrew,
Look at this pdf for the maths.

As its an heating element, the resistance of the element will be lower when its cold, so a higher current will flow while the element reaches its operating tempr/resistance.
 

Attachments

Last edited:
Hi Eric,

thanks for that.

Another question if I may. I could not find a suitable realy (off the shelf) that does 12V at 20A so I bought an automotive relay http://www.electro-tech-online.com/custompdfs/2009/09/CMA3XXX.pdf . The relay does not latch but is in working order.

The relay spec's quote a nominal voltage of 12V and a consumption of 1.6W. That gives me 133mA I need to drive out of the 2N3906 in order to make it latch. I don't think that there is enough output from the transistor but it should be able to handle 200mA so I will need to modify the circuit.

Reading up I think I need to alter the value of R4. I have been through a couple of tutorials but they are greek to me help please.

Regards
Andrew

PS: Still coming out in November? The Rand must be getting a little more expensive for you now.
 

ericgibbs

Well-Known Member
Most Helpful Member
hi,
The 2N3906 has a gain about 15 to 20 at 135mA, so you need to adjust the base drive resistor to give about 8 to 10mA minimum.

Its all set for Nov09, fingers crossed.:)
 
Hi

I tried a 4k7 resistor in place of the 10k. Noticed the 3906 was getting warm, don't know if as a result but think I did something before that. Put the 10k back, circuit back to original. The transistor switches on but only 1V travels to the relay, the tranny get warm. Replace the tranny, same result. Any idea of what I might have wrecked?

Andrew
 

tytower

Banned
with Hysteresis, whats happening is you are feeding back a little bit of voltage from the output when it is switched on . Doing so holds the input voltage just that bit higher than when it switched. The thing runs a bit longer to get the input voltage down to the set point again . At least thats how I see it at the moment.

I may be wrong so someone will correct me.

If that is correct above then reducing the feedback resistance will allow a higher voltage to feed back and the thing would run longer to get back to the set voltage. I did this with a freezer monitor ,it had a 10 Megohm feedback and I soldered another across it to reduce the overall resistance to 5 Megohm. The hysteresis gap increased a lot and I got rid of a chattering it had just at switching point. It randomly flashed an led in the line and actually sounded like it was picking up a radio transmission . So much so that I put a bench amp on it to check . It was just noise.
 
Last edited:

ericgibbs

Well-Known Member
Most Helpful Member
Hi

I tried a 4k7 resistor in place of the 10k. Noticed the 3906 was getting warm, don't know if as a result but think I did something before that. Put the 10k back, circuit back to original. The transistor switches on but only 1V travels to the relay, the tranny get warm. Replace the tranny, same result. Any idea of what I might have wrecked?

Andrew
hi,
The reason that the 3906 is getting warm is because its not saturating when switching 135mA, due to the low base current via the 10K.

For that transistor you require approx 10mA base current to ensure it saturates at 135mA.
This would mean a base resistor of only 1K, I suspect the 741 will give problems due to its very limited lower switching voltage with a single supply.

I would tackle the problem by choosing a transistor that has more gain at 135mA collector current, example would be a ZTX550, with say a 4k7 base resistor.
Changing the resistor will change the hysteresis levels, so the levels will need to be reset.
 
Last edited:
Hi Eric,

they did not have a ZTX550 but gave me a NTE129P. I popped this in along with a 4k7. The units switches on but not with enough power to energise the coil. The LED is very dim and the tranny get bloody hot real quick.

Any ideas.

Andrew
 

ericgibbs

Well-Known Member
Most Helpful Member
Hi Eric,

they did not have a ZTX550 but gave me a NTE129P. I popped this in along with a 4k7. The units switches on but not with enough power to energise the coil. The LED is very dim and the tranny get bloody hot real quick.

Any ideas.

Andrew
hi,
The NTE looks ok from the datasheet.

As I said earlier it maybe the old 741, can you measure the voltage on the output pin of the 741.?
When switched to power the relay and at rest.
 
Hi Eric,

Pin 6 Off 12.4V
Pin 6 On 2.2V

Sounds the wrong way around. Is there a better opamp with the same pinouts as the 741?

Cheers
Andrew
 

ericgibbs

Well-Known Member
Most Helpful Member
Hi Eric,

Pin 6 Off 12.4V
Pin 6 On 2.2V

Sounds the wrong way around. Is there a better opamp with the same pinouts as the 741?

Cheers
Andrew
hi,
As you can see the 2.2V low and the 1V or so drop on the NTE base/emitter adds upto 3.2V.
Take this from the 12V supply rail leaves about 9v across the 4k7 giving about 2mA base current, which is not enough to switch the NTE hard ON.

You could try a 1K, also remove the 10K that connects from the base to the 12V.

Can you try that first.?:)
 
Hi

Very impressive Mr Gibbs, it works a real treat.

Tried just with the 1k and left the 10k to +. No can do, popped out the 10k, LED is sparkling, relay is clicking strongly and I am a happy chappy.

Thanks a lot Eric
Cheers
Andrew

Now at least will have a cold fridge whilst en route.
 
Status
Not open for further replies.

Latest threads

EE World Online Articles

Loading
Top