Tektronix Technical Forums are maintained by community involvement. Feel free to post questions or respond to questions by other members. Should you require a time-sensitive answer, please contact your local Tektronix support center here.

6430 Output off Damage to DUT

6220, 6221 Current Sources
mdepew
Posts: 11
Joined: August 24th, 2015, 10:27 am
Country: United States

6430 Output off Damage to DUT

Post by mdepew » August 24th, 2015, 10:32 am

We are having an issue with the Keithley 6430 Current source. Out DUT has JFETs on its inputs, which are connected to the current source. When the current sources are off and the DUT is powered down, the triax cables are moved from one JFET to another. Sometimes when this change happens, the JFETs are damaged and begin to leak between their gate and source, sometimes to the mA level.

The output off mode is set to normal. Is there something about this configuration that could be inducing a shock into our DUT?

Also, we took similar actions with the 6220 current sources, but did not see damage. Is there a difference between the 6220 output off mode and the normal output off mode of the 6430?

Dale C
Keithley Applications
Keithley Applications
Posts: 2840
Joined: June 10th, 2010, 6:22 am
Country: United States
Contact:

Re: 6430 Output off Damage to DUT

Post by Dale C » August 24th, 2015, 11:07 am

Model 6220 or 6221 does not change the current setting when output is off.
Model 6430 in normal mode sets the output to 0V and current compliance to 0.5% of the present range.
And therefore should not damage any DUTS that are connected.
Can you try to use ZERO output off state instead of NORMAL?
Maybe that would help.

mdepew
Posts: 11
Joined: August 24th, 2015, 10:27 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by mdepew » August 25th, 2015, 7:13 am

So in the 6220, when it does not change the current setting for output off, does it output a 0A current, or does it open the circuit?

Is the Zero mode of the 6430 more similar to the 6220 output off mode, or is the Guard mode?

Thank you for the quick reply.

Dale C
Keithley Applications
Keithley Applications
Posts: 2840
Joined: June 10th, 2010, 6:22 am
Country: United States
Contact:

Re: 6430 Output off Damage to DUT

Post by Dale C » August 25th, 2015, 8:24 am

The normal mode of output off of the 6430 is the closest to the 622X output off state.
In the 6220 a relay open up for high impedance.
In the normal mode of the 6430 the V-Source is selected and set to
0V. Current compliance is set to 0.5% full scale of the present current range. In theory, with the
V-Source set to zero, the SourceMeter will not source or sink power.

mdepew
Posts: 11
Joined: August 24th, 2015, 10:27 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by mdepew » August 25th, 2015, 1:19 pm

We tried operating the 6430s in the Zero output off mode. We were then able to observe the current level when we move our JFET connections. We set the Isrc to 1pA, and we saw around 1pA of current flow out of one current source and 1pA of current into another. When the connection to a new jfet is made, we see the current reading go to overflow briefly before going to 1pA. In this mode we are still seeing damage to the JFETs, but it seems to be happening at a much slower rate. Is there a mode we can put the 6430 in where it will not drive ANY current?

Could we power down the 6430s entirely when we connect to a new JFET, and then power them back up after the connection is made?

Thank you.

brian.d.smith
Keithley Applications
Keithley Applications
Posts: 402
Joined: June 10th, 2013, 6:00 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by brian.d.smith » August 27th, 2015, 5:13 am

You may have have a problem with charge build-up in your device, or in the cables, or on your person. I would recommend placing a bleeder resistor across the current source output before you disconnect the FET and allow some time to bleed the charge.

Is your workstation outfitted with a static mat that is installed properly? Is the person performing these tests properly strapped into the static mat system?

mdepew
Posts: 11
Joined: August 24th, 2015, 10:27 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by mdepew » August 27th, 2015, 6:35 am

The person performing the testing is wearing a wrist strap and ESD coat, at an ESD certified workstation.

We have waited as much as an hour after power-down of our DUT to move the cables, which should allow ample time for any charge to drain off.

From what we can tell, it looks like one current source is driving into an other. Our setup includes two current sources, which are both connected to the drains of different JFETs. The sources of these JFETs are connected together, so when our DUT is powered off the outputs of the current sources are connected together.

When configured in Zero mode for output off, where we can see the current flow out of the 0V source, we see a positive current on one source and a negative current on the other source. It appears the current sources are driving against each other. The amount of current they can drive against each other appears limited in the steady-state case, but there is a current spike at the time of connection that goes off scale of the meter, so we don't know how big it is. Ideally we would like to connect these current sources while they are in a mode like the 6220 output off. Where no current can flow.

Would it be a problem to power down the 6430s entirely when cables are being moved and only power them up after our DUT is powered on? Would there be any concern about transients on the cables during the power-up process?

brian.d.smith
Keithley Applications
Keithley Applications
Posts: 402
Joined: June 10th, 2013, 6:00 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by brian.d.smith » August 27th, 2015, 7:49 am

Are you using two separate 6430 current sources?

Is there a power source in your setup other than the 6430s?

Can you make a drawing of the instruments, cables and your devices and post that here?

mdepew
Posts: 11
Joined: August 24th, 2015, 10:27 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by mdepew » August 27th, 2015, 9:36 am

Yes, we are using two separate sources. When our device is powered off, both JFETs are conducting. The control logic applies -4V to pinch one JFET off or the other.
6430 setup.png
6430 setup.png (27.75 KiB) Viewed 41421 times

mdepew
Posts: 11
Joined: August 24th, 2015, 10:27 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by mdepew » August 27th, 2015, 11:33 am

The JFETs are connected to the center conductor of the triax cable, the guard ring is unterminated on the DUT side, and the outer shield of the triax cable is connected to the DUT ground.

brian.d.smith
Keithley Applications
Keithley Applications
Posts: 402
Joined: June 10th, 2013, 6:00 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by brian.d.smith » August 30th, 2015, 9:22 pm

I am a bit confused. While the input to the op amp is at ground potential, in general, the current allowed into the op amp is very low. There would be no path to discharge the FET.

mdepew
Posts: 11
Joined: August 24th, 2015, 10:27 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by mdepew » August 31st, 2015, 2:04 pm

There is a feedback element on the op-amp that is not shown in the sketch.

brian.d.smith
Keithley Applications
Keithley Applications
Posts: 402
Joined: June 10th, 2013, 6:00 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by brian.d.smith » September 1st, 2015, 5:01 am

Have you tried changing operating mode of the 6430 to force 0V as a shutdown procedure? You could measure current and possibly observe the charge being removed from the FET.

Forcing current out of the 6430 makes the output extremely high impedance. It may be difficult to discharge the FET with the 6430 output in high impedance.

mdepew
Posts: 11
Joined: August 24th, 2015, 10:27 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by mdepew » September 1st, 2015, 7:56 am

We have tried setting the output off mode of the source to "Zero". In this mode the current sources have their outputs connected through the channels of the JFETs, and drive against each other. When we make the initial connection, there is a current spike, but the meter on the front of the source reads "overload" so we don't know the magnitude of the transient. We believe this transient is the source of the damage to our JFETs.

Is there a mode we can put the 6430 sources in where they will not drive against each other in output off mode? The relay in the 6220 sources effectively prevents this transient and we have not had any JFET damage issues with the 6220 sources.

brian.d.smith
Keithley Applications
Keithley Applications
Posts: 402
Joined: June 10th, 2013, 6:00 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by brian.d.smith » September 2nd, 2015, 7:54 am

My suggestion is to keep the output on/enabled after your test. When your current sourcing test is finished, change from current source to voltage source and set the voltage to zero and measure current. See if there is detectable current that is draining the charge and see if the current goes to zero. That should discharge the drain-source channel and allow you to perform your change of connections.

Even though your operator is wearing an ESD strap, the operator may induce a charge on your device just by being in proximity to the FET and moving around.

You should make sure the voltage of the gate is set to turn on the drain-source channel when you set the 6430 to voltage source mode and level to 0V. After discharging the drain you should discharge the gate with your gate supply.

Another thought is to place some banana connector near your FETs and connect them such that you can insert resistors to assist with the discharge of the FETs and keep the discharge resistors there during the cable change.

mdepew
Posts: 11
Joined: August 24th, 2015, 10:27 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by mdepew » September 2nd, 2015, 11:54 am

When the current sources are in zero voltage source mode, they never stop showing current flow because they are driving against each other up to whatever the compliance limit is set to. The JFETs have their drain-source channel turned on when the gate is grounded. When our DUT is powered off, the outputs of the two current sources are connected together through the two drain-source channels.

The issue seems to be when the 6430s first start driving against each other, there is a transient that exceeds the compliance limit. This transient appears to be damaging to our JFETs. We do not see similar damage when using the 6220s.

brian.d.smith
Keithley Applications
Keithley Applications
Posts: 402
Joined: June 10th, 2013, 6:00 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by brian.d.smith » September 2nd, 2015, 1:34 pm

What is the value of current that you are sourcing with each 6430?
What is the compliance voltage setting?

mdepew
Posts: 11
Joined: August 24th, 2015, 10:27 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by mdepew » September 2nd, 2015, 2:11 pm

The current source is set to 1pA. When the output is off, in the zero mode, that should mean that our compliance is 1pA.

We simplified the circuit and have been able to capture a few of the transients on a scope. (the resistors are a 1M to ground)
6430 setup.png
6430 setup.png (29.62 KiB) Viewed 41372 times
The bottom 6430 is connected first. When the top 6430 is connected we sometimes see a transient. The image below is of one that was 12V for ~1ms.
IMG_20150902_154611586.jpg
IMG_20150902_154611586.jpg (1.08 MiB) Viewed 41372 times

brian.d.smith
Keithley Applications
Keithley Applications
Posts: 402
Joined: June 10th, 2013, 6:00 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by brian.d.smith » September 3rd, 2015, 7:51 am

We set up a 6221 and then a 6430 into a high resistance resistor, set 1pA force and turned on and off the output. A 1M scope input impedance was used.

The 6430 produced about 70 mV pulse. The 6221 pulse was 1/3 of this or less.

So, the 6430 is less good at source settling for this arrangement than the 6221, but neither instrument produced a pulse as much as you are experiencing.

I am wondering if there is some coupling going on in your circuit that is sensitive to a source overshooting. Maybe the overshoot is being amplified by your FET?

Can you describe or show a - of how you are implementing a guard to prevent current flowing from HI to LO in the air? In general, at or below 1nA, noise and charge induced by proximity will swamp forcing or measuring current.

mdepew
Posts: 11
Joined: August 24th, 2015, 10:27 am
Country: United States

Re: 6430 Output off Damage to DUT

Post by mdepew » September 3rd, 2015, 9:45 am

The issue is not turning the sources on/off. The issue is connecting/disconnecting while the sources are in output off mode.

We configured the sources to zero mode, output off 1pA compliance, and that transient was produced at the time the 2nd 6430 contact with the JFET.

The JFETs were placed in a solderless breadboard. The gates were connected to the common ground through 1M resistors. The JFET sources were connected together and triax cables ending in clips were used to connect the low sides of both 6430s to the common ground. The high sides were connected one and then the other to the JFET drains. When the 2nd drain connection is made, the transient occurs.

Post Reply

Return to “Current Sources”

Who is online

Users browsing this forum: No registered users and 3 guests