Measuring DC Voltage

[KSJeff]

I'm having some issues with a PoE (power over ethernet) device that is sometimes failing to operate correctly. I have a suspicion that it's something on my switch. I've gone back and forth between the switch vendor and the phone vendor and neither is to blame (allegedly).

So, I need to find a way to measure and log DC voltages up to 50V. I could get by with a min/max button that would last 72 hours, but all seem to be short term. I've seen loggers as cheap as $50 for DC voltage, but it looks more complicated than I would like.

I figure I'm going to have to spend $200+ to do this easily or deal with cheap and complicated. But, I thought I'd ask to collective here. What would you do this with?

 

[Sevenhills1952]

How expensive is the unit? Maybe less expensive to just replace it? Or like the simple (single) $5 Walmart one just to troubleshoot by subbing in.
There are lots of multimeters available with min/max storage so you could leave connected, over time recall the voltage swing.

Sent from my SM-S320VL using Tapatalk

 

[nafterclifen]

I'm assuming that you've already verified compatibility in terms of...

1. Cable type (CAT5, CAT5e, CAT6, etc), including all terminations (jacks, connectors, couplers, etc)
2. PoE standard (15.4 W) vs PoE+ (25.5 W)
3. Total output capacity of switch (make sure you're not exceeding overall output)
4. Switchport output capacity

?

 

[KSJeff]

Yes, we're cat5e and we had a vendor on site that verified the links while they were running the cat6a. Full reports, not just continuity. Happening across multiple Meraki MS355 48 port switches with Avaya IP telephones. We put POE injectors on the phones and they run fine. Put them back on the Meraki and some of them don't. Didn't have any issues with previous 6500 series Cisco.

So, I want to see if it's dropping any DC voltage.

 

[KSJeff]

How expensive is the unit? Maybe less expensive to just replace it? Or like the simple (single) $5 Walmart one just to troubleshoot by subbing in.
There are lots of multimeters available with min/max storage so you could leave connected, over time recall the voltage swing.

They list at 8K, but I'm guessing we give less than 5K per 48 port switch. Right now I've got about 16 deployed and 10 pending.

I've got a fairly nice Wavetek multi, but it doesn't hold min/max that long. If I could find something that will run for three days without going to sleep that would produce min/max that would work.

 

[budmur]

Uni-T UT-61E connected to a laptop would do it, although you may need to power the meter from an external source.

 

[Falcon67]

Cisco is having some real issues these days, especially in the lower cost end switching. By lower end, I mean 5K or so and down. Our PoE WS-C2960X-48FPD-L switching runs around $3K - but we're .EDU so we get big discounts. If your Meraki switch has limited lifetime, get TAC on the line and demand a replacement. It's a PITA but if you're persistent they will replace.

"All Cisco Meraki MS switches include a limited lifetime hardware warranty that provides next-day advance hardware replacement. " Do it, quit screwing with it. Call TAC and open a case. If you already have an open case and the engineer is balking, jump their azz. Hammer it - you paid for it, it has a warrant, demand justice.

I've had that conversation over 8 times in the last 12 months on new switches that fail for all kinds of reasons - won't stack, brick after upgrade, report new IOS as counterfeit, system failures in general.

BTW, we use a $5K+ Fluke tester to verify PoE levels.

 

[KSJeff]

Unfortunately our Cisco phones work fine. I know we're getting PoE voltage, I just don't know if it ever fluctuates enough to cause problems with our Avaya phones. To add some confusion to the pot 90% of our Avaya phones are good. And they never completely lose power. I don't lose DHCP, and the workstation doesn't disconnect. Right now since I have a cable vendor on site, I'm having them test the lines, then I just replace the phone. But, my previous 6509s had no problems at all.

 

[nafterclifen]

The fact that the issue is happening with multiple switches and it goes away when you use a power injector, I agree with your suspicion of switch issues.

Unfortunately, I don't have any direct troubleshooting experience with your specific equipment so I can't offer any input. Other than the obvious - have you checked software/firmware for both the switches and phones?

And if you want to monitor a device (i.e. an IP) via ping to see if/when it stops responding and receive an email when it does, "Ping for life" is a simple free utility.

 

[MattT]

I know we're getting PoE voltage, I just don't know if it ever fluctuates enough to cause problems with our Avaya phones.

Dirty DC power is the other possibility. That can cause some very strange "random" behavior. A good DMM might show something suspicious with AC volts or frequency but a 'scope would be best way to check power quality. And if you get a USB Pico you can also use it for data logging.

https://www.picotech.com/library/data-loggers/picolog-6-data-logger-software

 

[Sevenhills1952]

Is the power supply plugged into an UPS system? That would insure a clean, steady 120vac output.

Sent from my SM-S320VL using Tapatalk

 

[MattT]

Is the power supply plugged into an UPS system? That would insure a clean, steady 120vac output.

A facility of that size should have clean AC feeding the switches. Though it wouldn't hurt to check that too.

 

[KSJeff]

Is the power supply plugged into an UPS system? That would insure a clean, steady 120vac output.

Yes, all the critical power in the facility is running from a 250kVA UPS. The switches are running at 220Volts. Just like the old ones.

I think I'm missing something on that Pico station. Does something in there record up to 50VDC? Look like temperature and current.

 

[MattT]

I think I'm missing something on that Pico station. Does something in there record up to 50VDC? Look like temperature and current.

You use the software with one of their oscilloscopes to log voltage. You'll have to do a little research to figure out whether it'll work with all their 'scopes or just some of them.

https://www.picotech.com/products/oscilloscope

 

[Sevenhills1952]

This any help?


Voltage and Current Data Loggers and Recorders
https://www.microdaq.com/data-loggers/voltage-current.php

Sent from my SM-S320VL using Tapatalk

 

[jeffmoss26]

Unfortunately our Cisco phones work fine.

Can I Still Call Out?

 

[Falcon67]

Unfortunately our Cisco phones work fine. I know we're getting PoE voltage, I just don't know if it ever fluctuates enough to cause problems with our Avaya phones. To add some confusion to the pot 90% of our Avaya phones are good. And they never completely lose power. I don't lose DHCP, and the workstation doesn't disconnect. Right now since I have a cable vendor on site, I'm having them test the lines, then I just replace the phone. But, my previous 6509s had no problems at all.

We use all PoE and USB phones now. My network is always perfect, my bandwidth is unlimited - therefore I'd blame the phones.

I have 2960X-48FPD switches installed that are supposed to be 370w PoE (as opposed to FDPs which are 740w). I have a couple that will not allow 24 803.3af APs to run, it has to be 23 or less. If I had a better budget, those would go in the trash. The APs pull less than 6w each. Part of the issue may be that some devices register as Class 0 which the IOS registers as needing 15.4w. Still, an LPD should run 24 af devices. What is also sad is that our old Dell 5524P units do a better job or reporting power used and complain less than the new model Cisco units. Check your "sho power inline" list and see what the switch says - it may "think" it's running low on power.

 

[dcg9381]

I think, in this circumstance, rather than try to log all the voltage, I'd take that particular device off POE and use a stand-alone POE injector (basically have it on it's own power source). If that solves it - we know it's something up with the source voltage..

 

[SGKent]

We had some issues a few years back that were intermittent and it turned out to be a bad florescent ballast injecting RF into the equipment. I happened to catch it start acting up one day at the exact same time the light flickered.

 

[todd_fuller]

You want something like this: https://www.microsemi.com/product-directory/poe-accessories/4801-poe-tester