REF1941: Capacitive load / Driving reference down a long cable

Hi Stephen,

How much do you currently have on the output of the REF1941?  The cable length does add capacitance to the output.  Have you tried changing the output capacitor to something smaller?

Ben

Hi Ben,

Thank you for the quick reply. It looks that this design is missing output bypass capacitors on the reference. On the next design 100nF will be added to both outputs. 

Regards,

Stephen

Hi Stephen,

Thanks for the feedback.  If there are no other questions, please close the thread by clicking on "resolved".

Ben

Hi Ben,

Would you suggest that, assuming the cable does not exceed the maximum stable output capacitance of 10uF, that the missing bypassing is the most likely culprit of the issue? The electrical characteristics in the datasheet state that the stable output capacitance is 0uF to 10uF.

Many thanks,

Stephen

Hi Stephen,

At the moment, there are no output capacitors on the board?  Before you add the cables that connect to the thermistors, check to see if the output of the REF1941 is stable and then add one thermistor cable at a time.  

Ben

Hi Ben,

No there is no output decoupling, but when the references were observed with a scope they looked stable, but since the issue seems intermittent there may be conditions with which instability is induced. I think what I will do is try and exaggerate the issue on the bench, and try to "force" instability by using an artificially long cable and seeing what effect this has on the reference. 

I am re-spinning the board as we speak and have added 100nF output bypassing, but I wondered if you had any other suggestions at this point as to any additional precautions I could take? Maybe adding a compensation resistor at the output?

Thanks

Hi Stephen,

If you are re-spinning the board, plan on adding additional landings for the capacitors on the output if space allows.

Just curious, do you have an idea of how much capacitance the cable has?  If you were to shorten the cable, do you see the instability?  

When you are observing the instability on some of the cables, what is load of the thermistor?  What if you were to add a dummy load to increase the output current, does the instability go away?

Ben

Hi Ben.

I've added the output caps. I am not yet sure on the cable capacitance of the cable. I have not yet been able to perform any tests on the bench. The load from the thermistors equate to 2.3 mA in total (at operating temperatures). I think until I am able to run some more tests and get more data I am unable to provide any more feedback.

Thank you for your help, I can either mark the post as resolved or I can wait until I have results from testing.

Cheers

Thanks Ben. Hope you have a great weekend too!