Comparator problem (basic low voltage cutoff)

[Skytrapdrums]

Hi,

I'm not completely new to electronics, I've done some PSU's and filters and such. But now I really feel stupid, because I can't seem to get a simple comparator circuit working, or the comparator working at all.

I've attached the circuit the I have simulated in Icircuit which of course is not a "real" simulation program, but I thought it might be of some use. The power supply in the circuit is meant to be between 10 an 14,5 Volts. The circuit is meant to be used to protect a LiFePo battery from running too low, and is supposed to cut the load at 10,5V.

After a few days of frustration I finally tried using two potentiometers on to inverting input and one to non-inverting input of the LM393N to just try to get the comparator to work but output stays at 0,55V at the maximum.

Can someone tell me what I'm doing wrong? Maybe I have managed to fry the chip?

 

[alec_t]

No way will that comparator drive the average relay directly. You are more or less shorting the output to ground (its spec says it should survive that). It can only source 16mA. You need a transistor buffer between comparator and relay coil.

Edit: Forgot the 393 has open-collector output. For 'source' read 'sink'.

 

[Skytrapdrums]

Ok thanks, I completely missed the maximum current, I've tested the circuit with a led but i think that one wants about 20mA and the circuit doesn't work. Still I tried just the comparator without load but the output doesn't seem to turn on when trying to give for example 2V to inverting input and 6V to non inverting. This probably means I've fried it. On the upside, their not expensive...

Does anyone know of a comparator that could manage about 200-500mA?


Thank you!

 

[ericgibbs]

Ok thanks, I completely missed the maximum current, I've tested the circuit with a led but i think that one wants about 20mA and the circuit doesn't work. Still I tried just the comparator without load but the output doesn't seem to turn on when trying to give for example 2V to inverting input and 6V to non inverting. This probably means I've fried it. On the upside, their not expensive...

Thank you!

hi,
Looking at your original circuit, you should note that the LM393 is an open collector output, the internal pin is just a collector, used as current sink , not a current source.

 

[alec_t]

Could you use a MOSFET instead of the relay? The LM393 would drive that ok.

 

[Skytrapdrums]

Yeah, I could use a MOSFET. I just don't know exactly which kind i should use. The application is basically three solenoids driving hydraulic valves rated at 30W 12V (battery as power supply) so i think I would need quite a big one. I do have some TIP147, TIP142's and MJE3055's but i can imagine that i might need a buffer transistor before driving the big ones. The problem is that I really don't have the experience of using transistors, I've mostly been playing with op-amps in filters and such. So just for starters I don't even know wether to use a PNP or NPN transistor.

And if any of you guys have an idea for a different approach without a comparator I'm all ears.

I've measured the the power drain of the solenoids and they really use only about 1 Amp when on.

I thank You for any and all input.

 

[ericgibbs]

hi
Look at the LM311 comparator, it has a higher current sink capability.

 

[alec_t]

The application is basically three solenoids driving hydraulic valves rated at 30W 12V

Are all three switched at the same time (i.e. sharing a common relay/transistor) or individually? We should be able to suggest suitable FETs.

 

[Skytrapdrums]

Well in the original schematic the relay is meant as the only channel for any power used from the battery. The same principle is for the FETs. But your question was actually meant if all valves are on at the same time (i guess) and this should not be the case, but from life so far I've learned the this possibility should be considered. So the 7,5 amps can be possible, but i read that the 3055 should be able to carry 15 amps.

I'm a bit amazed that you guys have the patience to help such a helpless case as myself, again thank you.

 

[alec_t]

The 3055 is a bit long in the tooth. It may well carry 15A but it will also drop a lot more voltage in doing so than a modern FET would, hence would get considerably hotter. If you go by the rule of thumb that the base current of a 3055 needs to be 1/10 the collector current to ensure transistor saturation then your comparator will need to drive 0.75A into the base for as long as the relay is on. That's a big ask. A FET would need only tens of milliamps of gate drive, at the instant of switching, and next to nothing at other times. Check online for 'NMOS 30A' or similar.