How to protect AO from transients?

[WhiteCloud]

Hi,

i have a hardware and electronic question, and hope anyone can help me with advice:

We are controlling the speed of a motor with a 0 to +10 V Analog Output Signal of a DAQ device. (The AO is connected with a motor controller). The controller is not very reliably and it happens that it regenerates voltage peaks of 60 V and transients which could destroy the Input of the DAQ.

The question: How can I protect the AO? My idea was to build in into the circuit a schottky-diode (Current only in direction of controller) and a TVS-diode which cuts all peaks. Please see the attached draft.

Would that be a good enough, or which is the usual stuff to solve this problem?

I also have a little problem to select a correct TVS diode, because I don´t understand what the characteristics like the

"Maximum reverse standoff voltage" and "Breakdown voltage" mean. So how should the characteristics of the diode be? (The signal is only from 0 to +10 V, so anything above theese limits can be eliminated with a shortcut...

Thanx all for your replies!!

[Henrik_Volkers]

here is a quick idea

R3, D2, V2 form the motor controller with a 60 spike, V1 is AO

Since you have a 0-10V output a simple zener would do the job. I choose a 12V type

R1 and R2 limit currents and protect the diode and the AO.

If your controller have a low input impedance the voltage drop over R1, R2  have to be regarded. A test would answer that question 🙂

However the current the AO sink still could be 12mA ...

Maybe the output is already clamped, and all what is needed is a current limiting resistor? That is a question only NI can answer.

I'm a fan of LTspice, simple and fast. Starting LV and LTspice at the same time, this circuit was ready while LV still was loading 😉 

(To be honest, the circuit yes, but the parameter where still missing)

[WhiteCloud]

Hi Henrik,

thanx for your reply! If I understood you correctly, the only component you would build in for protection is the zener-diode which shortcuts all negative spikes (Higher than 0,7 V) and all positive spikes above the +12 V level? R3, D2, V2 only simulates the spikes, is that correct?

You wrote:

"If your controller have a low input impedance the voltage drop over R1, R2  have to be regarded. A test would answer that question"

I am not an expert in electronics. How can i find that out? Our controller is a prototype...can that be simply tested with the "Ohm-function" of a multimeter? I know that the current which flows into the controller is not more than 10 mA...

What is the purpose of R1 and R2? Don´t they have an impact on my signal?

Thank you again, greetings Markus

[Henrik_Volkers]

R1 and R2 limit the currents in case of spikes.

R2 limit the current for the protecting zener to I(D1) = ( U(spike=V2) -U(zener=12V) ) / R2

R1 limit the current into the AO . Assume the worst case U(AO)=0V then the current it has to sink is ( U(zener)-U(AO) ) / R1= 12mA

If the input of the motor controller form a Ri= 10k input impedance you build a voltage divider

and the controller will see a voltage of Ui=U(AO) * Ri / (R1+R2+Ri)  

You apply 10V at the AO and the controller will only see 8.3V , however if the input impedance is 100k the controller will see 9.8V