Switch Hardware and Software

cancel
Showing results for 
Search instead for 
Did you mean: 

loud "zizi" sound: niSwitch DMM Switch Handshaking.vi PXI 4071/PXI 2527

Solved!
Go to solution

I am running the handshaking example: "niSwitch DMM Switch Handshaking.vi" for DMM PXI-4071/SWITCH PXI 2527 in a PXI-1033 Chassis to do a test measurement of a DC 1 Volt.

 

When I hit "Run", the switch block makes continuous loud "zizi" noise; the "measurements" graph gives simply a 0 volt line. This is obviously wrong results.

 

My setting for this VI is as follows:

 

handshaking.jpg

 

Does anybody know what is wrong here?

 

Thank you,

 

Bing

0 Kudos
Message 1 of 9
(7,170 Views)

Your scan list only contains one channel in it.  Thus, you're repeatedly measuring channel 0 to com.  I'll assume you've connected the DMM to COM± and have a DUT connected to channel 0±.  If you're DUT isn't outputting 0V, then we should disconnect your DUT from channel 0 and connect a AA battery (etc) across the CH0± leads.  Now when you run, you should repeatedly measure ~1.2V.

 

Next, if you want to measure more than one channel, you'll need to add the other channels to your list.  With a multiplexer, a shortcut to measure sequential channels is "ch0:31->com0;"  This will scan channels 0 through 31 sequentially.  If you only want to measure channels 0 through 7 and 10 through 15, you could use "ch0:7->com0; ch10:15->com0;".  If you only want to hook up individual channels - say 1, 5, 7, 13 - you could use "ch1->com0; ch5->com0; ch7->com0; ch13->com0;" etc.  The entire syntax is here:

http://zone.ni.com/reference/en-XX/help/375472C-01/switch/scan_list/

It's also on your computer in the NI Switches Help document, located in the NI-Switch subfolder of the National Instruments folder in your start menu.

 

Your triggers are configured correctly, so from the sound of it (you are hearing a sound, so that's reassuring) it's just the scan list that needs updating.

 

Happy switching!

-John Sullivan
Problem Solver
0 Kudos
Message 2 of 9
(7,154 Views)

Hi John, 

 

Thank you for your help! I just found I made a mistake that I didn't connect DMM to com. My device is working now and the measurement gives also right results.

 

But the continuous annoying "zizi" noise persists, when I run a programm which scans channels. I can tell this "zizi" sound comes from the Switch. It sounds like something is scratching somewhere.  I don't know if this sound is normal or something wrong is happening. I am a bit worried that this "zizi" may hurt the Switch. 

 

I have attached the recording of the "zizi" sound from my Switch.(please change the extension of the attached files: .jpg-->.wav , beause .wav is not a valid file extension for this forum).

 

Thank you,

 

BIng

Download All
0 Kudos
Message 3 of 9
(7,143 Views)
Solution
Accepted by topic author liubingxy

Blng,

 

This noise is actually the sound of the relay actuating. Since these relays are electromechanical relays, when the relay closes, the two contacts essentially slap against each other and produce this noise. The continuous buzzing that you hear is because the relays are opening and closing repeatedly. If you use the device in NI Switch Soft Front Panel for example, you can close each relay individually and hear a single click per relay that is actuated.

Frank,
National Instruments
Software Group Manager
0 Kudos
Message 4 of 9
(7,132 Views)

Hi Frank,

From your explanation, it seems the Switch is working correctly and nothing is wrong there. I just hope this repeat open and close won't affect the lifetime of the Switch much.

Thank you,

Bing

0 Kudos
Message 5 of 9
(7,127 Views)

Bing,

 

The repeating of the open and close will degrade the lifetime of the relay. If you need to measure more than one channel, scanning is a good option as John mentioned, and the example "Switch Scanning with DMM - Handshaking.vi" is a good option. If the intent is to only measure a single channel then scanning is not required and you could create a VI (or even use the soft front panel) that closes a single relay and then takes as many measurements as needed before opening the relay.

Frank,
National Instruments
Software Group Manager
0 Kudos
Message 6 of 9
(7,121 Views)

Hey Bing,

 

If you pull up the 2527 specifications, you can see the mechanical and purely resistive relay lifetimes.  For example, if you are only mechanically switching a relay (i.e. there's no electrical signal going through that relay), then you could expect on the order of 100 million cycles before the relay fails.  Note that as soon as you start switching a signal, the relay lifetime will drop substantially due to the additional contact wear from the electrical arc.  For example, with a purely resistive load and less than 30mV across the relay and less than 10mA through the relay, you could expect on the order of 2.5 million cycles on any particular relay before failing.  Another example is 10000 cycles switching a resistive 30VDC load at 2A.

 

You can see that the voltage and current across and through the relay while switching exponentially decreases expected relay lifetime.  These numbers are general guidelines; certain relays could last an order of magnitude longer.  Also, as soon as you start switching capacitive or inductive loads, the reactance is going to greatly reduce relay life.

 

So yes, your concerns about relay lifetime are valid: all mechanical relays will eventually fail, but with proper estimation you can predict how long a particular relay should last in your test system. 

-John Sullivan
Problem Solver
0 Kudos
Message 7 of 9
(7,115 Views)

 

Hi John,

 

Thank you for your detailed explanation. I think you have made this issue quite clear to me.

 

For my measurement, I need to use 8 channels at one time and this 2527 Switch has 32 channels. So I will use ch0:7 first; then ch8:15 and ....ch24:31. Does this mean the lifetime of the Switch used in this way is four times of a single relay's?

 

Bing

0 Kudos
Message 8 of 9
(7,105 Views)

Hey Bing,

 

In the 2-wire 32x1 case, yes, each relay is essentially mutually exclusive from all others.  Not all topologies are this way; sometimes, there's a single relay that muxes either all even or all odd channels, but that is not the case on the 2527 in 2-wire 32x1.  So yes, once you've reached end of life on channels 0 through 7, you could switch to 8 to 15, and start from scratch. 

 

FYI: We list the relay topology in our specification sheets and in the NI Switches Help.  For the 2527, it's figure 2 in the spec sheet.

-John Sullivan
Problem Solver
0 Kudos
Message 9 of 9
(7,098 Views)