Hello guys! First time here!

I am currently working on a power supply for a medical application where I have to charge an NiMH 12V battery when the voltage across it is lower than a certain point. So, for that matter, I am using a non-inverting Schmitt trigger, that is to make sure to charge only when necessary and bringing the battery back to its nominal point. The problem is, I am not sure if the LM317 constant current source (200mA) will negatively affect (that is fry!) the LM358 when it is not charging the battery (when the battery GND is not connected). I ordered the PCB, but now that it's on the way, I am having a little doubt over it. Could you guys tell me if I messed up on that part?



Thanks to all of you,

Jeff

 

I don't see a problem when everything is at operating voltage. Worst case current could be about 3 ma. input with the 5.1K input resistor, which shouldn't be a problem.

Although, the opamp could be damaged from static from battery connector. You might think about some anti-static protection on that line.

 

The output voltage of the LM317 can never exceed the 18V supply voltage, and in fact will always be less by the dropout plus the drop across D3. So I don't see how that could ever put the LM358 at risk.

You might want to protect your regulators by placing a diode from output to input. This ensures the output cannot be at a higher voltage and cause damaging reverse current to flow.

 

Thanks for your help! Sometimes, when obsessed by what could go wrong, we tend to forget some basic principles! @ Lestraveled: I will certainly think about that antistatic protection for a future version.

 

It is an important part of the design process to go back and look for danger points after you have finished getting the circuit to function correctly. Double-check the power ratings on your resistors, protect the inputs against static discharge, visualize the start-up sequence when power is first applied, make sure your copper traces are large enough for the current they carry, make sure none of the transistors will glitch shorted when power is removed, etc. Nothing OCD about that.

 

I'm not so sure you don't have a problem. The max input voltage is Vcc - 1.5 volts as I read the data sheet. But you can fix it with a rail to rail op amp.

Edit:
After looking at the schematic, I think it just won't work above Vcc - 1.5, but no damage.

 

I'm not so sure you don't have a problem. The max input voltage is Vcc - 1.5 volts as I read the data sheet. But you can fix it with a rail to rail op amp.

Edit:
After looking at the schematic, I think it just won't work above Vcc - 1.5, but no damage.

Well, I only need it to compare the voltage in the range between 11.5V and 13.5V, so there is no problem in there! Thanks for taking a look!

 

You will only see the 16-18 volts if the battery is disconnected. You are using that op-amp as a comparator anyway. You will probably get lucky on this one. If not, ronvs suggestion of a rail to rail op-amp will solve it.

 

You will only see the 16-18 volts if the battery is disconnected. You are using that op-amp as a comparator anyway. You will probably get lucky on this one. If not, ronvs suggestion of a rail to rail op-amp will solve it.

Yeah...I sure hope to get lucky! I sure do understand the problem, but I tested this part of the circuit with 18V on the input of the LM358 during maybe 5 minutes and everything seemed to work just fine WITHOUT the Vcc plugged in. I only had a 1.7V output at the gate of the MOSFET in that case (nothing big enough to switch it ON). So...I guess I will be fine! Thanks for all the insight!