Electronic Load Prototype Test Fail

Thread Starter

Roessleralec

Joined Jul 16, 2019
23
My current project is an electronic load. To test it before I design a pcb, I decided to build a very simple prototype. The circuit is your common load circuit thats all over the internet with an op amp, mosfet, and sense resistor.

1575060920651.png


The op amp is a tle2141 and the mosfet is ixth64n10L2. However after running a test, I got a weird voltage, definitely not the expected 5v at the inverting input of op amp. I wanna say it was around .6v. Now I thought this was weird so I built another circuit using the common LM358 and IRFZ44n. I used this because many loads have been successfully built with these components. This time the inverting input was .56v. My understanding of the circuit is that the op amp will drive its output to whatever value it takes the get its inputs to the same voltage. This means that there should be 5v at the resistor and according to ohms law 500ma flowing through the circuit.

I went into LTSPICE and simulated that exact circuit, only to get 2.5mv.


1575061474696.png


Does anyone have any idea of whats going on?
 

bertus

Joined Apr 5, 2008
20,189
Hello,

The supply voltage of 5 Volts for the opamp is not enough to drive the mosfet.
Raise the voltage to 15 Volts and see what happens.

Bertus
 

crutschow

Joined Mar 14, 2008
23,808
My understanding of the circuit is that the op amp will drive its output to whatever value it takes the get its inputs to the same voltage.
Yes, that's true within the opamp's operating limits.
But the opamps you are using both do not tolerate input voltages up to the supply voltage (see below for the input range of the TLE2141 with a 5V supply).
The LM358 is similar.
You need a rail-rail type op amp for the input and output to go to the supply rail voltage.

Alternately increase the supply voltage as bertus suggested.

1575062516439.png
 

Thread Starter

Roessleralec

Joined Jul 16, 2019
23
Yes, that's true within the opamp's operating limits.
But the opamps you are using both do not tolerate input voltages up to the supply voltage (see below for the input range of the TLE2141 with a 5V supply).
The LM358 is similar.
You need a rail-rail type op amp for the input and output to go to the supply rail voltage.

Alternately increase the supply voltage as bertus suggested.

View attachment 193426
I replaced the op amps 5v supply with 19v and now get .5v at the inverting input. Changing the supply voltage in LtSpice now gives me 22mv. It is now well above the 5v reference and still within its specs.
 

Papabravo

Joined Feb 24, 2006
12,702
You still don't understand the problem. You cannot take the input of the opamp closer to the supply rail than about 2 Volts or the part will malfunction. That is the meaning of "Common Mode Input Range". The same restriction applies to both inputs
 

Thread Starter

Roessleralec

Joined Jul 16, 2019
23
You still don't understand the problem. You cannot take the input of the opamp closer to the supply rail than about 2 Volts or the part will malfunction. That is the meaning of "Common Mode Input Range". The same restriction applies to both inputs

Yes I understand. The non-inverting input is 19-5 = 14v difference and so the inverting input should also be. This way they are both within spec.
 

Papabravo

Joined Feb 24, 2006
12,702
Yes I understand. The non-inverting input is 19-5 = 14v difference and so the inverting input should also be. This way they are both within spec.
OK I guess I referred back to the original schematic and did not appreciate all of the changes. So what opamp model is being used for U1 now?
 

Thread Starter

Roessleralec

Joined Jul 16, 2019
23
OK I guess I referred back to the original schematic and did not appreciate all of the changes. So what opamp model is being used for U1 now?
Yes sorry if I came off wrong. The LtSpice simulation was primarily to make sure I had done everything right and it wasn’t a mistake in my connections. Therefore I just used the universal op amp model for simplicity. In the real circuit I am currently using an LM358 and plan to eventually use a tle2141.
 

Papabravo

Joined Feb 24, 2006
12,702
Yes sorry if I came off wrong. The LtSpice simulation was primarily to make sure I had done everything right and it wasn’t a mistake in my connections. Therefore I just used the universal op amp model for simplicity. In the real circuit I am currently using an LM358 and plan to eventually use a tle2141.
No we're cool (or at least cooler than the opamp). Still trying to understand the problem.
 

Thread Starter

Roessleralec

Joined Jul 16, 2019
23
I have the circuit that you see in my first post (LtSpice schematic) with irfz44n as the mosfet and lm358 as the op amp. I also changed the supply voltage to 19v to be within the op amps specs regarding input vs supply difference. My problem is that instead of seeing 5v at the inverting pin(and resistor/mosfet source) I am seeing ~.5v. If it helps the output voltage is relatively constant at 4v. After leaving this circuit running for a few minutes the op amp was extremely hot. Hope that clears it up.

BTW thanks for the replies
 

Alec_t

Joined Sep 17, 2013
10,611
Have you checked the datasheet to ensure the pinout of the FET is correct? It sounds as though the opamp is trying to drive the FET substrate diode.
 

Thread Starter

Roessleralec

Joined Jul 16, 2019
23
The circuit is built according to this Pinout. To write off a possibility, I swapped the source and drain wires and the resistor almost blew up. It got really hot and started fizzling. I think I can conclude that my Pinout is right.
Also I have included a picture of my testing circuit. The blue wire is 19v in and the orange connected to the power rails is 5v.
 

Attachments

RamaD

Joined Dec 4, 2009
328
If 500mA is flowing, assuming that the opamp outputs 5V, Vgs will be zero. The MOSFET will not conduct!

If you are aiming at 500mA, then reduce the resistance to 1 Ohm. Reduce the voltage at the positive input of the opamp to 0.5V. Use an opamp with rail-to-rail output, or use LM358 with higher Vcc, say +12V.

With full 500mA flowing, the voltage across the resistor will be 0.5V. Opamp output will be Vgs+0.5V.
 

TeeKay6

Joined Apr 20, 2019
532
I have the circuit that you see in my first post (LtSpice schematic) with irfz44n as the mosfet and lm358 as the op amp. I also changed the supply voltage to 19v to be within the op amps specs regarding input vs supply difference. My problem is that instead of seeing 5v at the inverting pin(and resistor/mosfet source) I am seeing ~.5v. If it helps the output voltage is relatively constant at 4v. After leaving this circuit running for a few minutes the op amp was extremely hot. Hope that clears it up.

BTW thanks for the replies
There is a "wire" connecting the opamp inv input to the MOSFET source. As I understand you, one end of the wire (opamp inv in) is at 0.5V while the other end of the wire (MOSFET source) is at 4V. Doesn't that seem strange to you? It tells me that your schematic does not represent your circuit, or that you have damaged components/connections, or possibly missing/wrong grounds.
 

volkwyn

Joined Dec 10, 2019
2
Is your maximum load voltage V2 = 19 V?
What is the maximum load current you want to set?

Some of my comments are in line with comments already posted here.

The IRFZ44N is optimized for Vgs of 10V. With an IRFZ44N, the LM358 should have a supply voltage of 12V or higher.

If you really would like to use a +5V supply for the op amp, then use a logic level MOSFET which would turn fully on at Vgs=2.4V or thereabouts.

Change R1 to 1 ohm, and then 0.5 V input to the op amp will give 0.5 V across the sense resistor R1 and give a load current of 0.5 A.

A Mosfet has a very high input resistance, and can sometimes be turned on by touching the gate lead or wire to the gate with one’s finger. A series resistor of say 1k ohm in the gate circuit ie between op amp output and gate may prevent such accidental turn on when you are working with the op amp wiring.

There is a "wire" connecting the opamp inv input to the MOSFET source. As I understand you, one end of the wire (opamp inv in) is at 0.5V while the other end of the wire (MOSFET source) is at 4V. Doesn't that seem strange to you? It tells me that your schematic does not represent your circuit, or that you have damaged components/connections, or possibly missing/wrong grounds.
Definitely, if there is more than a few millivolts across an op amp’s inverting and non inverting inputs, there has to be a wiring/ground fault or damaged components
 
Top