Keithley 2400 SourceMeter Quick Test (Help Needed!)

[Hugoneus]

Dear members,

Does any of you have a Keithley 2400 SourceMeter handy?

If so, can you do quick output short measurements where you directly short the output terminals together without using a cable? What resistance does it report to you? (You have to make a very low-resistance short of course).

I would very much appreciate this.

[mickpah]

 Cf
D

[Hugoneus]

Cf
D

Hmmm... That is not as helpful as I thought it would be! 

[free_electron]

What are you after ? Simply turning on ohms meter function uses regular two wire sensing with these machines. You need to go in the menu and explicitly turn on 4 wire sensing. Then you can measure low stuff.

[Hugoneus]

What are you after ? Simply turning on ohms meter function uses regular two wire sensing with these machines. You need to go in the menu and explicitly turn on 4 wire sensing. Then you can measure low stuff.

The problem is as follows:

I set the instrument to 1A and connect a precision 1-Ohm resistor at the output. I use the Measure-V function and it reports 1.3V. I put the multimeter across the unit's output terminals (right at the instrument) and I read 1.01V.

I would like to know if this is normal for this instrument to have an internal resistance of 0.3-Ohms... It seems rather high.

[free_electron]

No. Thats not normal... Let me doublecheck later today. But that sound very high. That stub of wire between the board and connectors is barely 1 inch long... Could be no more than 50 milliohms.

It is possible you have transient resistance in the connector. Ive seen that....

[Hugoneus]

No. Thats not normal... Let me doublecheck later today. But that sound very high. That stub of wire between the board and connectors is barely 1 inch long... Could be no more than 50 milliohms.

It is possible you have transient resistance in the connector. Ive seen that....

I got a hold of another Keithley 2400 and it has the same problem. It seems to be a design issue. 0.3-Ohm of internal resistance is very high. So in another words, if you are working with currents above 10mA, use the sense ports or it will give you the wrong voltage readout.

What I find troubling is that the internal voltage readout is suppose to be at the terminals. But it clearly is not!

[free_electron]

There is only onewire going from the main board to the red terminal and one wire for the black terminal. tghe banan jacks are NOT set up as kelvin connections in these machines !

That's why they have sense inputs when doin any kind of precision work you need kelvin wiring.

[Hugoneus]

There is only onewire going from the main board to the red terminal and one wire for the black terminal. tghe banan jacks are NOT set up as kelvin connections in these machines !

That's why they have sense inputs when doin any kind of precision work you need kelvin wiring.

I opened the unit and measured the voltage on the PCB itself! And the voltage still does not match the readout. Therefore, there must be a 0.3-Ohm resistance somewhere between where the voltage is sampled on the PCB and where the wire is soldered to the PCB. Not good.

[SeanB]

Connect source and sense connections together and try again.

[Hugoneus]

Connect source and sense connections together and try again.

Everything works with the sense. But the point is that the internal 0.3-Ohm resistance is far too high. Furthermore, the reported voltage is not of the terminals but of some internal node. No other power supply in my lab has such a poor output resistance.

[Hugoneus]

I contacted Keithley. They have re-produced the problem and are investigating it. I will let you know what they say.

[TiN]

I can try on my 2400 later, if you still need this.   

Happy 2013, btw