Hi everyone!,
first at all I need to say sorry for my english, Im not anglo speaker.
Im working doing massives curves IV, I have a array of 6 diodes so I switch manual the connection to measure and then I change the array for a new array.
So, I thought to implement a mux with a arduino to control via python the outs and let it run alone.
The problem is the signal OUT isnt the same that IN.

I have read this post: https://forum.allaboutcircuits.com/threads/problems-with-using-mux.79599/

And now I know that the impedance is important to the result but I dont how to implement it.
Im using this mux: 74HC4067 (I bought it with the arduino so Its implemented in a arduino board)


Here is a example of the failing IV curves:

(FAIL)

(Correct)
So my question I think is easy for a person that know this knowledge, I need a mux that dont fail my measures. Im good at programming but with hardware Im a bit potato..
Thanks to all and best regards!

 

Perhaps the lines are not the same because sometimes we measure at room temperature and others in cryo, the problem is the elbow that occurs in the curve. Another issue that I forgot to mention is that we also measure in Bdv and we reach voltages that can exceed 50v.

 

The fact that there is a sharp knee in all the curves at exactly 1V suggests to me that it may be due to a software issue rather than the mux hardware.

we also measure in Bdv and we reach voltages that can exceed 50v.

Hopefully you're not applying 50V directly to the mux inputs .

 

The fact that there is a sharp knee in all the curves at exactly 1V suggests to me that it may be due to a software issue rather than the mux hardware.

Hopefully you're not applying 50V directly to the mux inputs .

Hi Alec!
Its imposible a software error, I only select the channel to measure with software. I have test it with a led board and the mux works but at measure IV curve We can see that knee. That should be some leakage current or some impendance that I'm not taking into account.

I applied 50v to the diode, and then I measure the current with a electrometer in serie, the mux make 6 to 1 in the channel select.

 

The fact that there is a sharp knee in all the curves at exactly 1V suggests to me that it may be due to a software issue rather than the mux hardware.

Hopefully you're not applying 50V directly to the mux inputs .

Wait, I have read your answer again and maybe you are wrong, the knee at 1v its normal, Its the direct polarization of the diode, the problem is the knee at 1.5v in the upper picture (Down picture its correct, manually select without mux)

 

Your mux has a 60Ω typical on-resistance, which gives a 600mV IR drop at 10mA.

You either need to use a mux with a lower on-resistance, or use two muxes, one to switch the applied voltage and one to read the diode voltage.
Since the device reading the voltage has a relatively high impedance, the mux on-resistance will not significantly affect the value read.

But you can't read the 50V reverse voltage value with that mux.
It's only good to its supply voltage.

 

Your mux has a 60Ω typical on-resistance, which gives a 600mV IR drop at 10mA.

You either need to use a mux with a lower on-resistance, or use two muxes, one to switch the applied voltage and one to read the diode voltage.
Since the device reading the voltage has a relatively high impedance, the mux on-resistance will not significantly affect the value read.

But you can't read the 50V reverse voltage value with that mux.
It's only good to its supply voltage.

OOh nice, that the answer I was looking for, so I can read reverse bias voltaje but no direct. I need a mux with lower than 60ohm.. to read direct. Thanks a lot sir!

 

Two Sets 10 channels O'scope and Multimeter Multi-Inputs Selector.

 

OOh nice, that the answer I was looking for, so I can read reverse bias voltaje but no direct. I need a mux with lower than 60ohm.. to read direct. Thanks a lot sir!

Your mux has a 60Ω typical on-resistance, which gives a 600mV IR drop at 10mA.

You either need to use a mux with a lower on-resistance, or use two muxes, one to switch the applied voltage and one to read the diode voltage.
Since the device reading the voltage has a relatively high impedance, the mux on-resistance will not significantly affect the value read.

But you can't read the 50V reverse voltage value with that mux.
It's only good to its supply voltage.

Can you explain me better how to use 2 mux? I dont understand how can I use both. Thanks in adavnce.

 

Two Sets 10 channels O'scope and Multimeter Multi-Inputs Selector.

I will study this answer but look a lot dificult for that purpose. Thanks!

 

I will study this answer but look a lot dificult for that purpose. Thanks!

To see the square diagram is easier to understand how it works.

 

To see the square diagram is easier to understand how it works.

Thanks a lot, I will try and I will notice you with news.

 

Can you explain me better how to use 2 mux?

Simple.
The first mux supplies the voltage/current to the devices with its 16 I/Os connected to the diodes.
The other mux's 16 I/Os are also connected to diodes to read the diode voltage, with its common I/O going to the voltage measurement device.
Both muxes are connected to the same 4-bit address control signal, of course.

Make sense?

 

Sorry for my 5 years old picture, but something like that? Yeah, I can implemet it, but I dont understand why.

 

View attachment 262306
Sorry for my 5 years old picture, but something like that? Yeah, I can implemet it, but I dont understand why.

Its mean eliminate the parallel at the start that supply the diodes no?

 

something like that?

No.
My scheme is to measure the forward voltage curve shown in your first post.

So the 16 I/O's of the two muxes are connected together with the second common I/O going to the voltage measurement instrument.
This separates the measurement circuit from the voltage supply circuit.

All the diode cathodes are connected to ground.

 

A detector of 40 pins connector, for 89C2051,89C2041,8951,8952, pic, arduino, etc...

As for the above circuit, you can change it to using the 74HC4067*2 and CD4051*1.

 

No.
My scheme is to measure the forward voltage curve shown in your first post.

So the 16 I/O's of the two muxes are connected together with the second common I/O going to the voltage measurement instrument.
This separates the measurement circuit from the voltage supply circuit.

All the diode cathodes are connected to ground.

Ahhh I understand totally now, I will try tomorrow and notice you as soon as posible.

 

Good morning @crutschow Im trying your solution but now Im thinking in a problem, with your solution, Im not measuring current in parallel? I will post another diagram

Maybe I continue without understand but your last explanation sound me to that.