In message 1444483218108.6a07c91c@Nodemailer, "Chris Waldrup" writes:

Then don't.

Instead get 12 or 24 Volt sealed lead-acid batteries and a good
float-charger, and run your stuff from that.

You avoid a lot of conversion losses, and you get to decide what
quality batteries you want (As opposed to "the cheapest we can get
away with") and you get to decide how long hold-up time you want.

The important tricks are:

1. ATO Fuse RIGHT NEXT TO THE BATTERY.  Not a meter away, but
   quite literally bolted right onto the terminal.

2. Don't buy a shit charger, it will cost you battery life.

3. Suitably sized fuse/polyfuses on all loads.

4. Either put 0.010 Ohm current shunts in all over the place
   or buy a 1mA resolution clamp meter and prepare the wiring
   for measurement.

And that's it really...

I run all the always-on stuff in my lab from two 12V/105Ah telco-grade
sealed lead-acid batteries, and I'll never look back.

Presenty the load is 6.7A @ 24V, and that powers my ADSL lines,
firewalls (soekris), home server (ITX with mini-box.com PSU),
emergency lights (LED strips), GPS, GPSDO, HP5065 etc. etc.

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

Hi

All of the UPS vendors these days make a wide range of products. They range from
low cost to mighty expensive. They also range a bit in terms of performance. Finding out
exactly what this or that model does do can be a major pain. The marketing guys apparently
don’t want you to figure out that the low cost gizmo does not have all the performance
of the one that costs 5 times as much.

First thing to avoid - the pure square wave output versions. These may or may not
generate RFI in this or that UHF band. They will take out big chunks of HF and mess up 10 MHz
distribution. I suspect that the filters on some are better than the filters on others. The bigger issue
is that they do not play nice with modern power factor corrected power supplies. These supplies
seem to expect a sine wave and a (possibly ringing) square wave may get them confused.

Next thing to avoid - The stepped square wave / modified square wave versions. Theses aren’t
quite as bad as the centuries old square wave units. It’s more likely you will find these than a
pure square. They still have the same issues on RFI. They may or may not antagonize a PFC
supply to the same extent.

The target is a “pure sine wave” output. That keeps a PFC supply happy. As with any verbal spec,
“pure” likely has a few qualifiers on it. It’s not guaranteed to take out the RFI. It at least will reduce
the spikes on your 1 pps lines. If you look at the output on a ’scope the waveform at least looks like
a sine wave.

All of these gizmos run a switcher in the “few hundred KHz” range to generate the output. They run
a similar switcher to charge the battery. None of them are compatible with an indoor antenna in the same
room trying to listen to MF. None of the ones I have tried are nasty enough to bother GPS, either indoor or
outdoor. Put another way, I’ve had more trouble from LED lights than from sine wave UPS’s.

Cyber Power (via Amazon) seems to make some pretty good stuff at a price point a bit below APC. As
with everybody else, it’s a pain to figure out what’s what. Are you after a “big” rack mount unit (as in $500
and up) or something smaller?

Bob

Hi over the last 10 years or so I've purchased several consumer / small office grade UPS's from sources such as Staples and Costco.  I've never noted any RFI from them but I live in a (rf) noisy urban neighbourhood so any noise they put out is probably hard to notice.

Other than occasionally turning some of them off and on and seeing if I can see or hear any (additional) RFI on frequencies of interest I've never really looked for RFI from them.

Several years ago I started putting ferrite chokes on the feed lines for my antennas and have almost completely  switched to double shielded cables for my amateur radio and time nuts activities.

Hope these comments are of some interest.

Mark S
VE7AFZ

Sent from my iPhone

A quick comment about "pure sine" devices.

At times I operate a portable VHF and up radio system from rf quiet out of the way places.  I run almost all the equipment from lead acid storage batteries.  I have small 30 dollar inverter I use to occasionally  power some equipment that needs 120 volts AC.  It puts out very little RFI.  I figured I'd upgrade to a much more expensive pure sine Inverter (I also wanted to run some linear power supplies from the inverter as well to get better voltage regulation for some of the equipment.)  It was a disaster from an RFI perspective.

Your mileage may vary.

Mark S
VE7AFZ

Sent from my iPad

On 2015-10-10, at 7:51 AM, Bob Camp kb8tq@n1k.org wrote:

Hi,

Thanks for the comments this morning. 

I'm not looking for something huge, maybe in the under $200 range. Just enough for the Thunderbolt, laptop, and Agilent 53131A counter. I'd like a little headroom though, as I may add another Thunderbolt or a rubidium standard someday. 

I do like the 12 V battery idea. I run my ham gear off a 12 V gel cell (I'm mainly QRP so a 32 Ah cell works for me) and I have a 12 V distribution system with Anderson Power Poles on my operating bench. 

I do a lot of VLF and LF listening but my antennas are outside.  Nothing at this time above 2 meters. 

 I live in a very quiet rural area on top of a mountain in Tennessee. So not much man made noise at all and I wanted to keep it quiet emission wise. 

Thanks for all your help.

Chris

KD4PBJ

—
Sent from Mailbox

On Sat, Oct 10, 2015 at 10:05 AM, Mark Spencer mark@alignedsolutions.com
wrote:

for a very similar application I am using a solar panel to charge the
battery, but I have a vented NiFe battery, which is not sensitive of
over charging or deep discharging, and has almost unlimited life time --
I have seen some in forklifts which were 60 years old and working...
73
KJ6UHN
Alex

On 10/10/2015 7:14 AM, Poul-Henning Kamp wrote:

Hi

If you do any VLF, then anything with an inverter in it is going to be a potential disaster. It
will purely be a case of the inverter being at a frequency that bugs you right now. None of
the ones I have seen are tightly controlled in frequency. What they do today probably will not
be what they do tomorrow.

Indoors vs outdoors does not matter as much when a tenth wavelength is into the next county.
The inverter in one of these gizmos is effectively a 1KW transmitter at (say) 60 KHz. Yes filters
can help. They will rarely do the trick when you are trying to listen to something at 60, 120, 180,
240 or 360 KHz. Since that’s 60 KHz +/- lots, those “wiped out” segments are also pretty wide.

Bob

Chris Waldrup kirjoitti:

Get a line-interactive model which has inverter with classic iron
transformer. Line-interactive means that the inverter is not
continuously on when the mains voltage is OK, but the mais voltage is
routed thru multitap transformer which gives some filtering and enables
to fix under/overvoltage errors without turning on the inverter. When
there's blackout and inverter is really started the transformer will act
as a filter which reduces the HF interference caused by sine wave inverter.

You can recognize this kind of ups from treir weight. For example 2 kVA
model should weight more than 25 kg without batteries and more than 50
kg with batteries. You could look for example old APC SmartUPSes or APC
Matrix UPS, which has separate inverter unit, transformer unit and
battery units. Old SmartUPS'es may require float voltage modification
because their charging voltage tends to increase when aging. Without fix
it will kill the batteries too soon.

Worst case to buy is double conversion on-line model without any kind of
transformer. These are cheap but the inverter is on all the time and
it's output is not filtered. I have measured the output of these kind of
UPSes with spectrum analyzer and they are really horrible interference
sources. Not recommended if there's any kind of sensitive electronics.

Regards,

--
73s!
Esa
OH4KJU

I have 3 APC SmartUPS2200NET UPSs.  I have detected no interference to
my HF ham station from these.  One antenna is several hundred feet away;
another passes less than 10 feet away.  I have listened to these with an
IC-R10, and not found much noise.

I get much more from noise radiated by ethernet cables, unless I have
choked them (which is effective).

I bought a sun-power (I think) 1KW sine wave inverter for ham radio
field day use, to run antenna rotors, etc.  While not a disaster, it
will have to be in a shielded box and all cables in and out choked, to
not interfere with HF communications.  Based on rough measurements, I
don't think much of a problem at 1 GHz.

Good luck!
Jim
wb4gcs@amsat.org

On 10/10/2015 2:32 PM, Esa Heikkinen wrote:

This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

On 10 October 2015 at 14:20, Chris Waldrup kd4pbj@gmail.com wrote:

There's one issue with them that I don't see anyone mention.

I was thinking of doing the same a while back, and intended getting a UPS
and adding a large external battery pack, so if the mains failed late at
night, I could run the GPS receiver and a few other things overnight, and
consider starting the generator in the morning.  I contacted a dealer on
eBay, who specilaises in UPSs. He told me that the smaller units with built
in batteries will die if you put large external batteries on them.
Essentially the charging circuits are not designed to run as long as needed
to charge big batteries. Even on ones designed for external batteries,
there's a recommended limit on the size of them. So if you think you might
want to increase runtime by adding some batteries, buy one designed for
that service.

I've had two here which were HP/Compaq 5 kW units. These were different to
the normal, in that the batteries added up to over 300 V, so could produce
240 VAC with no need to step it up. Both these blew up on me, for reasons I
never worked out. The load was never anywhere near 5 kW.

Lots of people mention sine wave. Of course, if you keen enough, you could
make a class A amplifier and sine wave oscillator. The problem is that the
pure sine wave inverters tend to be very inefficient.

As with most things, there are a lot of things to balance - runtime, cost,
quality of output, audio noise, RFI  etc etc.

Dave

sine wave UPS device.

I acquired a Liebert 2KVA UPS for data centers in 1999 when our division
was declared redundant. The cost when new would be prohibitive, but this
one was junked when it stopped working. I got it, found a bad
electrolytic cap (from the stains) and replaced it. Worked well. Never
did get a manual, had to trace the switching stuff.

This was from the days when data centers could afford the best in
immortal power. The line feeds a regulated DC supply that runs the sine
wave inverter and maintains battery charge. The inverter runs in sync
with the line, but a relay connects the line to the load while the
inverter idles. When the line fails, batteries supply the inverter and
the relay switches it to the load. Then a cooling fan starts up.

This machine required eight 12 volt batteries for 96 VDC nominal, higher
on float charge. I bought 15 amp-hour batteries and wired them to a
connector on the back intended for external batteries. Lethal voltage,
of course. Requires proper protection.

Also had a line voltage relay and toggle switch that would bypass the
UPS and connect the time rack directly to the line. This allowed the UPS
to be replaced when its warranty ran out. Batteries lasted six years,
did not invest another $500 in them.

Good luck.

Bill Hawkins

-----Original Message-----
From: Chris Waldrup
Sent: Saturday, October 10, 2015 8:20 AM
To: TIme Nuts
Subject: [time-nuts] UPS for my time rack

Hi,

I have decided I'd like to get a UPS to put on the rack containing my
Thunderbolt, the laptop that runs Lady Heather, and frequency counter.
Has anyone had bad experience noise wise with the APC brand units like
are available on Amazon and at Staples? I'd like to get one that doesn't
generate lots of RFI. Thank you. 

Chris
KD4PBJ

Just to add to this.  I've also been told that other pure sine inverters have worked well for others in similar applications.

Not sure if mine was unusually bad or the frequencies it operated on happened to be related to the frequencies I was using.

Your success may vary.

Best regards
Mark S

Sent from my iPhone

On 10/11/2015 12:07 AM, Dr. David Kirkby (Kirkby Microwave Ltd) wrote:

I have gone down that route, so I have some real data to share:

My (soon to be replaced) backup is an old back-ups CS 500, with a
rewired battery pack out of an RT3000 UPS. So instead of 7Ah, the UPS
has 40Ah. With plenty of fuses.

When charging the standard 7Ah battery, the UPS delivers about 0.7A
(from memory) for many hours, and sits at about 14C above ambient.

When charging the 40Ah, the current is the same, the temperature is the
same, just for longer, as it should be, since the thermal time constant
is much shorter than the time it takes to charge the 7Ah.

Where this has problems is during discharge:
I have about 55W load on it, which in turn is at least 5A on the
battery. After 2 hours a timer in the UPS shuts it off, regardless of
battery voltage.
Also, if you run the UPS at high load where the standard battery lasts
shorter than the thermal time constant, then there might well be trouble.

The replacement, a back-ups pro 1500 behaves differently.
It has support for external battery packs, and will happily run for at
least 5 hours, even when the external-pack-present signal is not connected.
The external-pack-present signal does make a difference when charging;
Without it, it charges at 0.7A. with it, it charges at 1.5A, and the fan
is on continuously.

/Kasper Pedersen

Hi

Unless you live in an unusual location, long term power outages are going to be pretty
rare. At the house I’m now in, we had a high voltage feed that was on it’s last legs. We
had short outages on a “many times a week” basis if the wind was blowing at all. We had
rare outages in the > 5 minute range. The short / frequent blip stuff is what most light weight UPS’s
are designed to handle. Not everybody has this problem. I no longer have it, they
ripped out 10 miles of old feeder and the new one works fine.

Indeed there are locations that experience multi hour outages on a fairly regular basis.
The combination of bars closing late on Saturday and a long straight road with an abrupt
turn in it was particularly hard on a feed line I once had to cope with. In that case gas turbine
generators were the answer.

If you have a case where long outages are common, rotary machines are often the better
answer than batteries. In the case above, the power company was the one footing the bill
for the gear. Fair in this case since they were the ones that could have moved the line.

If a > 10 minute outage is a “less than once a year” sort of thing, and OCXO’s are your only concern,
let them shut down. The net impact to your lab will be relatively small. The cost to fix the problem will
be relatively large. Short blips often, are well worth fixing.

The hidden issue with running a UPS is the relatively short life of the batteries.
Sealed lead acid is low cost up front, but they simply do not last when charged the way
a typical UPS charges them. Before you go into “can’t be true” mode … plug a 100W light
bulb  load into your UPS and see how long it runs. Battery still shows as good on the indicator.
Gizmo only runs for 1/4 the time it should … hmmmm …. It’s very common to go into these
projects with a  reasonable budget, and then find out that the budget to keep it going is
not quite so generous.

Bob

Good morning all,

On 10/10/2015 5:07 PM, Dr. David Kirkby (Kirkby Microwave Ltd) wrote:

Years ago, I used to work at a place that would give me all their
cast-off computer parts, including old UPS'.  I've taken probably opened
up about 20 of them, from various manufacturers.

The problem is that most don't have good heat-sinking on the power
transistors.  One unit of a brand I can't remember, all it had for a
heat sink was an aluminum bar!  Clearly, it was only designed to be run
for a small length of time.

(It's the same case for the transformers - on several removed
transformers, I've tried to use them backwards to get 12 VAC from 120
VAC, and they overheat quickly at more than 50% load.)

If you add external batteries, likely you will overheat and fry the
electronics as they get far hotter than they ever intended.

thanks much and 73,
ben, kd5byb

Where I live, there are two problems.  Frequent long outages. Solved
with a natural gas standby generator, which has run several times in
anger for extended periods since installed.  (Vulnerable supply, low
priority for restoration.)

The bigger problem is transients.  On a good night, my computer UPS
activates at least once an hour.  SOmetimes you can see the lights
blink, some times not.

In the last 6 months, I have had problems with a UPS (recovered by
extended shutdown), an Astron Power supply for amateur radio equipment,
a spectrum analyzer, two signal generators, a network analyzer, and an
oscilloscope.  All were power supply failures, not all repairable.  My
lab is now protected from the power company by a SmartUPS 2200NET.

I expect the grid to get /less/ reliable.
-- In 2007, DOE published a grid study which said there did not
exist sufficient generation capacity over load to maintain grid
stability.  Insufficient additional generation was booked for
construction, so they predicted widespread rotating blackouts by 2010.
-- The 2008 recession greatly suppressed aggregate load, which is
probably why the rotating blackouts did not happen.  I read recently
that demand has not yet recovered above the suppressed levels following
the precipitous drop in 2008.  (Which generates interesting other
questions . . . .)
-- I have lost track of the number of GigaWatts of generation which
has been shut down.

If the load ever recovers . . ..

Good luck!

Jim
wb4gcs@amsast.org

On 10/11/2015 9:05 AM, Bob Camp wrote:

This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Hi

If your problem is transients, from lousy power companies or from lighting on your power line, there are ways to address that.
High voltage at the service line into the building should be fixed at the point the line comes in. If you don’t, then you get into
all sorts of neat “transient went here, then there, then nuked the gizmo”. For things like time or frequency distribution this
is a very real thing. The power to one set of gear may / may not be on the same UPS as the power to another set of gear.

The best answer is a whole house protection device. These can range from < $100 at your big box store to > $1K. They can be
a wire it in on a spare breaker or a call out a pro sort of thing. A lot depends on just how bad your problem is. The net effect is that
your line is fully clamped to local ground at the panel. The phases are both clamped to a level that should not affect properly
designed gear that’s in good condition. There is always a tradeoff between how tight to clamp and how fast the gizmo wears out.
There is the usual  “get what you pay for” in terms of knowing the current degree of wear out of the device.

The alternative is to get your entire lab (and all the devices that feed it) onto a good isolation transformer. . Everything then ties to
a single “lab ground”. What ever bounce you get is now all on everything at once. Each time I’ve done this, keeping all the standard
lines, antenna feeds, ethernet cables, GPIB cables, cable TV feeds, and the rest of it correct has become impossible after a year or two.
There’s just to much going all over the place.

We are already more than just a bit  off topic for this list. There is another twist this can take, that heads over to talking to
your power company and the people who regulate it. I have seen that work (as in the nice new line that feeds this side of town).

Bob

Dave,
You could use a 120V relay and switch the high capacity battery from its own
charger to the battery pack in the UPS.  When power comes back, the relay
automatically switches the battery out and back to its own charger.

Bob

UPS

night, I

consider

who

batteries

needed to

there's a

service.

produce

reasons I

could

the

cost,

Did the transient suppressors, too.

A few years ago in a severe winter storm, after we got the neighbor's
house on generator, his furnace still wouldn't work -- furnace brain
fried.  Took the repair guy 4 hours to make 20 minute trip . . . . my
surge suppressors went in a week later.

One thing I left out of my earlier post:  ANYTHING I care about is on a UPS.

Good luck!

Jim

On 10/11/2015 12:24 PM, Bob Camp wrote:

This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus