HELP - VOLTAGE TO CURRENT

[N3ur0n15]

In order to control a Solenoid Proportional Valve with a signal ranging 4mA - 20mA depending on a 1V to 5V signal, I implemented the circuit shown below. Turns out it has been working perfectly for about 6 months. All of a sudden, it just generates 8mA with ANY input voltage and a 220ohm resistor. I am using a IC LM324 and Mosfet IRF640N for Q5 and IRF4905 as Q6 in image.

I have implemented the circuit again, I have changed and seen if the ICs are working perfect, I have varied the source power from 12V to 18V. Everything to me seems fine. Then again, it only generates 8mA. I seriously don't know what can be the issue. Thanks in advance.

 

[dknguyen]

Rather than poke around blindly, measure the voltage at each opamp input and output, and transisor gate and source moving left to right to see if they are what they should be to narrow down the issue.

 

[N3ur0n15]

Thanks, I already have. They seems not to be showing the corresponding voltages as expected from simulation. I have bought new ICs and it's just the same. I don't know what can be causing the malfunction.

 

[dknguyen]

Thanks, I already have. They seems not to be showing the corresponding voltages as expected from simulation. I have bought new ICs and it's just the same. I don't know what can be causing the malfunction.

The point was to narrow down the problem area. So why are you still looking at the entire circuit if none of it reads correctly? If none of it reads correctly, then something is amiss near the beginning. So check every component there and compare to the simulation.

What is this circuit being built on? There was no chance for physical damage or corrosion was there?

 

[N3ur0n15]

The circuit is on a PCB fiberglass. I then replicated the circuit on protoboard to check it. There has been no corrosion or physical damage.

 

[dknguyen]

Why is one power pin on each op amp not connected?

 

[MichaelaJoy]

Have you checked the MOSFETs?

 

[alec_t]

Where are you measuring 8mA?
Have you checked the power supply?

 

[N3ur0n15]

MichaelaJoy Yes, I have.
alec_t At the circuit's output: the diode.

 

[dknguyen]

Have you tested the output with a resistor? Right now just the zero ohm ammeter is the load.

 

[N3ur0n15]

Yes, the valves I'm using have 220ohm aprox. So, my test resistor is of 220 ohm.

 

[alec_t]

If the circuit drives a solenoid, a catching diode across the coil would be needed.
Simulation shows that an inductive load may cause the circuit to oscillate. This might confuse your DMM.
What is the purpose of D3?

 

[Beau Schwabe]

I would re-check your circuit design... In my simulator I only see a variance of 27mA to 34mA.
I think R13, R14, R16, and Q5 could be eliminated the way you currently have the circuit drawn.
Instead connect pins 1 & 2 of U2:A and also connect that same node to pin 12 of U2: D

For that matter you could eliminate U2:A and C7 as well and just have your input going to pin 12 of U2: D
otherwise you just have two voltage followers/buffers ... no need for two.

 

[N3ur0n15]

alec_t inductive load where? The purpose for D3 is to carry the 4mA to 20mA signal out as an input for the valve.
Beau Schwabe The voltage source for the LM324 is 12V, the voltage signal that varies the current signal is 1V to 5V. Expecting 4mA at 1V and 20mA at 5V. The circuits works both in simulation and in real life. The case is that is just stopped working and I haven't found the cause. Thanks for the advice on R13,R14,R15 and Q5, nice.

 

[alec_t]

@alec_t inductive load where? The purpose for D3 is to carry the 4mA to 20mA signal out as an input for the valve.

I assumed the solenoid coil to be the inductive load. But I realise now the valve input is not the coil. So scrub my reference to the catching diode.
The P-MOSFET current is DC, so D3 isn't needed.

 

[N3ur0n15]

alec_t Oh, ok. I misunderstood you there. But still, I don't think D3 is causing the problem.

 

[N3ur0n15]

alec_t About that catching diode. Where would I put it? I'm sorry I'm an undergrad student and am not familiar with this one.

 

[alec_t]

I edited post #15.
If the valve input is 220\( \Omega \), ground tied as per your ammeter in post #1, couldn't you simplify the circuit to this :-

 

[N3ur0n15]

alec_t Thanks for the clarifications on post #15. I will simulate your diagram. The valve's resistance is around 220Ω.

 

[KeepItSimpleStupid]

Some general comments:

No decoupling caps.
U2:3 needs a weak pull-down
The gates of Q5 and Q6 need weak pull downs.

it's helpul when troubleshooting to look at the difference of the inverting and non-inverting inputs. In MOST designs, it should be close to zero