Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Jeff Lelko]

Hi All,

So I recently added an UPS to my FOH rack.  The unit is an Eaton PW9130L1000R-XL2U and has so far worked perfectly without a single glitch.  Being a new unit, I wanted to let it run for a few days before sending it out on a job just to be sure it won't decide to randomly shut down or power cycle, so to check this I plugged a standard nightstand alarm clock into the unit, figuring that if I ever come home to find the clock flashing I'll know the UPS failed.  Well, that never happened, but the oddity is that the clock ran fast.  It'd gain about 15 minutes every 12 hours.  I of course metered the power (120v, 60Hz), and repeated the test with the clock plugged into the same circuit upstream of the UPS without the symptom occurring.  So my questions is what is the UPS doing to cause the clock to run fast, yet every other piece of equipment plugged into it runs fine (2x Denon DN-501C, A&H Qu-Pac, BBE DS24, Motu MIDI Express XT, Powered USB Hub)?

Thanks!
-Jeff

[Daniel Levi]

The only thing I could think of is if the clock is referenced to the 60hz mains frequency (like in "analogue" clocks with synchronous motors) and whilst you might be reading that it has 60hz the measurement device might be slightly off. Everything else is more than likely quartz or crystal controlled and so isn't affected.

[Corey Scogin]

Hi All,

So I recently added an UPS to my FOH rack.  The unit is an Eaton PW9130L1000R-XL2U and has so far worked perfectly without a single glitch.  Being a new unit, I wanted to let it run for a few days before sending it out on a job just to be sure it won't decide to randomly shut down or power cycle, so to check this I plugged a standard nightstand alarm clock into the unit, figuring that if I ever come home to find the clock flashing I'll know the UPS failed.  Well, that never happened, but the oddity is that the clock ran fast.  It'd gain about 15 minutes every 12 hours.  I of course metered the power (120v, 60Hz), and repeated the test with the clock plugged into the same circuit upstream of the UPS without the symptom occurring.  So my questions is what is the UPS doing to cause the clock to run fast, yet every other piece of equipment plugged into it runs fine (2x Denon DN-501C, A&H Qu-Pac, BBE DS24, Motu MIDI Express XT, Powered USB Hub)?

Thanks!
-Jeff

Do you have a scope to check out how pure the output sine wave is? I wonder if it has high frequency switching noise and maybe the clock isn't well protected from such noise or RF.

[Jeff Lelko]

The only thing I could think of is if the clock is referenced to the 60hz mains frequency (like in "analogue" clocks with synchronous motors) and whilst you might be reading that it has 60hz the measurement device might be slightly off. Everything else is more than likely quartz or crystal controlled and so isn't affected.

That's what I was thinking along the lines of.  Once I get the rig back into the shop I'll try to use my better meter to check the frequency - the initial measurement was done with a basic Kill-a-Watt P3. 

Do you have a scope to check out how pure the output sine wave is? I wonder if it has high frequency switching noise and maybe the clock isn't well protected from such noise or RF.

I don't unfortunately.  I might try to borrow one if this is worth really digging into.  I suppose the important thing is that all the 'real' equipment works, but like most of us I get nervous anytime I think something's not right with electrical equipment.  Being an online double-conversion unit, I thought the output would have been better/cleaner than the input, and maybe it is, but I'm still a bit unsettled that there might be a deeper concern to address.     

[Mike Sokol]

Being an online double-conversion unit, I thought the output would have been better/cleaner than the input, and maybe it is, but I'm still a bit unsettled that there might be a deeper concern to address.     

Because it's an online double-conversion UPS, there's no phase lock between the input and the output. On some of the older double-conversion UPS units I used to install for big computer installations, there was a little potentiometer you could adjust to zero in on 60 Hz. However, even if you do set it as close as you can to 60 Hz there will likely be drift that will allow it to change from 59.999 to 60.001 Hz over the course of a day. And just a small fraction of a percent off of 60 Hz is enough to make a digital clock drift. Oh yes, Hammond B3 organs too...

[David Buckley]

Because it's an online double-conversion UPS, there's no phase lock between the input and the output.

That's very unusual as an out-of-the-box setting, though many UPSs have the ability to operate the inverter to an internal frequency source.

The reason its unusual is that a UPS cant provide enough current to clear faults, usually the best they can do is 150% for a few cycles, which with a short wont clear a breaker, so the UPS output would just collapse.  So what happens is when the UPS gets overloaded it switches back to street power, to provide the fault clearing current, and then a couple of seconds later switches back to inverter.

The switching is done with semiconductor switching, invariably thyristors.  But for this to work, the inverter output needs to be in phase with the input, so the UPS will phase track the incoming mains.

[Scott Holtzman]

That's very unusual as an out-of-the-box setting, though many UPSs have the ability to operate the inverter to an internal frequency source.

The reason its unusual is that a UPS cant provide enough current to clear faults, usually the best they can do is 150% for a few cycles, which with a short wont clear a breaker, so the UPS output would just collapse.  So what happens is when the UPS gets overloaded it switches back to street power, to provide the fault clearing current, and then a couple of seconds later switches back to inverter.

The switching is done with semiconductor switching, invariably thyristors.  But for this to work, the inverter output needs to be in phase with the input, so the UPS will phase track the incoming mains.

The Lorain Power ConstAC's used a PLL and it was slaved to the mains.  In free running mode (mains failure) it had a stability rating in PPM's, I forgot the stability and used a temperature controlled high stability oscillator.  When the mains were online it was within 1/2 phase if I recall.  They were not thyristors though,  SCR's that I called "choppers" made the most god awful buzz when running.

We just ordered a new Eaton 9390 loaded with 60KVA of rectifier and inverter capacity.  It's scalable to twice the capacity.  It's a dual bus system too.  I have to come up to speed on it so I will let you know how the phase lock works. 

I have an old Ferrups and it reads both input and output frequency so this concurs with Mike's statement.

[David Buckley]

They were not thyristors though,  SCR's that I called "choppers" made the most god awful buzz when running.

The actual devices that do the inverting are usually IGBTs, or in older designs, MOSFETs.  The static switch (sometimes called a transfer switch) is invariably SCRs, a/k/a thyristors.

[Scott Holtzman]

The actual devices that do the inverting are usually IGBTs, or in older designs, MOSFETs.  The static switch (sometimes called a transfer switch) is invariably SCRs, a/k/a thyristors.

What are the DV/DT ratings of those IGBT packages?  I thought that once you got above 10kva you were out of that type of package and into SCR's or triac's 

[Jonathan Johnson]

The only thing I could think of is if the clock is referenced to the 60hz mains frequency (like in "analogue" clocks with synchronous motors) and whilst you might be reading that it has 60hz the measurement device might be slightly off. Everything else is more than likely quartz or crystal controlled and so isn't affected.

I can't speak to the UPS (Mike Sokol explained it pretty good above), but digital clock radios plugged into AC mains almost exclusively are regulated by the mains frequency.

Sony digital clocks (and maybe others) have an internal battery backup. When mains power is lost, they switch to an oscillator circuit (not necessarily quartz) to regulate, and that circuit is horribly variable.

As long as your equipment is not highly dependent on synchronous mains frequency, you're probably OK.

You could have problems if you feed a digital signal between two devices that are on separate power sources as their internal word clocks will not be synchronous. (This can be an issue with DC powered digital devices too.)

Digital devices on the same power source will probably be OK, as the frequency will be consistent between them.

Since many devices us an internal oscillator rather than mains frequency for regulation anyway, mains frequency becomes irrelevant.

Trying to synchronize a digitally recorded track with digitally recorded video is subject to the same issues even with perfect mains frequency. Since without word clock sync each device uses its own oscillator, the attempt to overdub can be a real challenge!

[ProSoundWeb Community]