Accuracy of utility voltage reported by UPS

classic Classic list List threaded Threaded
10 messages Options
Reply | Threaded
Open this post in threaded view
|

Accuracy of utility voltage reported by UPS

Jan Ceuleers
Dear list,

Please take a look at the following table which was generated by multimon.cgi:

APCUPSD UPS Network Monitor
Tue May 17 18:44:05 CEST 2016
System Model Status Battery Chg Utility UPS Load UPS Temp Batt. Run Time Data
skr03 network Back-UPS ES 550G ONLINE
100.0 % 238.0 VAC 13.0 % - 29.9 min. All data
fe3 media Back-UPS ES 550G ONLINE
100.0 % 234.0 VAC 14.0 % - 28.2 min. All data
zotac server Back-UPS CS 650 ONLINE
100.0 % 236.0 VAC 32.0 % 29.2° C 23.4 min. All data

The supply to my house is single-phased, so the voltage seen by all three UPSes should be the same, except for ohmic losses in the in-house wiring (which I would estimate to be quite small due to the low current) and potential voltage variations over time (which are possible though unlikely to be the only explanation).

The 4V difference between the first two entries is particularly surprising because these are the exact same type of UPS, purchased at the same time, have a very similar load and are actually connected to the same power outlet, so that ohmic losses can be completely excluded.

I realise that these are consumer-grade UPSes, but are they supposed to be this bad at measuring utility voltage?

Thanks, Jan

------------------------------------------------------------------------------
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
_______________________________________________
Apcupsd-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/apcupsd-users
Reply | Threaded
Open this post in threaded view
|

Re: Accuracy of utility voltage reported by UPS

Trevor Roydhouse
Jan Ceuleers wrote:
> The 4V difference between the first two entries is particularly
> surprising because these are the exact same type of UPS, purchased at
> the same time, have a very similar load and are actually connected to
> the same power outlet, so that ohmic losses can be completely excluded.
>
> I realise that these are consumer-grade UPSes, but are they supposed to
> be this bad at measuring utility voltage?

Bear in mind that the line voltage is being sampled at different times,
depending on when each instance of apcupsd was started.

That said, my four SU1500I Smart-UPS disagree as to when the line
voltage should be trimmed/boosted despite having the same sensitivity
settings (though different firmware and manufacturing dates spread over
6 years).

------------------------------------------------------------------------------
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
_______________________________________________
Apcupsd-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/apcupsd-users
Reply | Threaded
Open this post in threaded view
|

Re: Accuracy of utility voltage reported by UPS

William P.N. Smith
In reply to this post by Jan Ceuleers
Without a good DVM to measure the exact voltage it's hard to be sure, but are you expecting better than 1% accuracy from a cheap UPS?

William P.N. Smith
ComputerSmiths Consulting, Inc.

On May 17, 2016, at 1:44 PM, Jan Ceuleers <[hidden email]> wrote:

Dear list,

Please take a look at the following table which was generated by multimon.cgi:

APCUPSD UPS Network Monitor
Tue May 17 18:44:05 CEST 2016
System Model Status Battery Chg Utility UPS Load UPS Temp Batt. Run Time Data
skr03 network Back-UPS ES 550G ONLINE
100.0 % 238.0 VAC 13.0 % - 29.9 min. All data
fe3 media Back-UPS ES 550G ONLINE
100.0 % 234.0 VAC 14.0 % - 28.2 min. All data
zotac server Back-UPS CS 650 ONLINE
100.0 % 236.0 VAC 32.0 % 29.2° C 23.4 min. All data

The supply to my house is single-phased, so the voltage seen by all three UPSes should be the same, except for ohmic losses in the in-house wiring (which I would estimate to be quite small due to the low current) and potential voltage variations over time (which are possible though unlikely to be the only explanation).

The 4V difference between the first two entries is particularly surprising because these are the exact same type of UPS, purchased at the same time, have a very similar load and are actually connected to the same power outlet, so that ohmic losses can be completely excluded.

I realise that these are consumer-grade UPSes, but are they supposed to be this bad at measuring utility voltage?

Thanks, Jan
------------------------------------------------------------------------------
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
_______________________________________________
Apcupsd-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/apcupsd-users

------------------------------------------------------------------------------
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
_______________________________________________
Apcupsd-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/apcupsd-users
Reply | Threaded
Open this post in threaded view
|

Re: Accuracy of utility voltage reported by UPS

bill
On 5/18/2016 6:38 AM, William P.N. Smith wrote:
Without a good DVM to measure the exact voltage it's hard to be sure, but are you expecting better than 1% accuracy from a cheap UPS?
If they are reporting to the precision of a tenth of a degree, it is reasonable to expect at least that precision.
-- 
Bill Drescher
william {at} TechServSys {dot} com

------------------------------------------------------------------------------
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
_______________________________________________
Apcupsd-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/apcupsd-users
Reply | Threaded
Open this post in threaded view
|

Re: Accuracy of utility voltage reported by UPS

Adam Płaszczyca

But realize the moment of measurements aren't same. Voltage of supply is varying few volts and it is normal, so if each device is picking voltage reading with several second delays between it could be difference between.

18 maj 2016 13:46 "bill" <[hidden email]> napisał(a):
On 5/18/2016 6:38 AM, William P.N. Smith wrote:
Without a good DVM to measure the exact voltage it's hard to be sure, but are you expecting better than 1% accuracy from a cheap UPS?
If they are reporting to the precision of a tenth of a degree, it is reasonable to expect at least that precision.
-- 
Bill Drescher
william {at} TechServSys {dot} com

------------------------------------------------------------------------------
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
_______________________________________________
Apcupsd-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/apcupsd-users


------------------------------------------------------------------------------
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
_______________________________________________
Apcupsd-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/apcupsd-users
Reply | Threaded
Open this post in threaded view
|

Re: Accuracy of utility voltage reported by UPS

William P.N. Smith
In reply to this post by bill
While you and I might understand repeatability, accuracy, and precision, without checking the specs I wouldn't expect too much from a cheap UPS.

I'd start with taking repeated measurements from all three boxes and analyze the numbers.

I've seen too many instances where 2 degrees C is reported as 35.6F (where did all those additional significant figures and extra accuracy come from?) to believe that a $60 UPS can read to 1/10 of a volt.

William P.N. Smith
ComputerSmiths Consulting, Inc.

> On May 18, 2016, at 7:19 AM, bill <[hidden email]> wrote:
>
>> On 5/18/2016 6:38 AM, William P.N. Smith wrote:
>> Without a good DVM to measure the exact voltage it's hard to be sure, but are you expecting better than 1% accuracy from a cheap UPS?
> If they are reporting to the precision of a tenth of a degree, it is reasonable to expect at least that precision.
> --
> Bill Drescher
> william {at} TechServSys {dot} com
> ------------------------------------------------------------------------------
> Mobile security can be enabling, not merely restricting. Employees who
> bring their own devices (BYOD) to work are irked by the imposition of MDM
> restrictions. Mobile Device Manager Plus allows you to control only the
> apps on BYO-devices by containerizing them, leaving personal data untouched!
> https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
> _______________________________________________
> Apcupsd-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/apcupsd-users

------------------------------------------------------------------------------
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
_______________________________________________
Apcupsd-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/apcupsd-users
Reply | Threaded
Open this post in threaded view
|

Re: Accuracy of utility voltage reported by UPS

William P.N. Smith
And, as Adam points out, the AC supply isn't stable from cycle to cycle (much less however the UPS is measuring it) to 1/10 volt.

William P.N. Smith
ComputerSmiths Consulting, Inc.

> On May 18, 2016, at 9:08 AM, William P.N. Smith <[hidden email]> wrote:
>
> While you and I might understand repeatability, accuracy, and precision, without checking the specs I wouldn't expect too much from a cheap UPS.
>
> I'd start with taking repeated measurements from all three boxes and analyze the numbers.
>
> I've seen too many instances where 2 degrees C is reported as 35.6F (where did all those additional significant figures and extra accuracy come from?) to believe that a $60 UPS can read to 1/10 of a volt.
>
> William P.N. Smith
> ComputerSmiths Consulting, Inc.
>
>>> On May 18, 2016, at 7:19 AM, bill <[hidden email]> wrote:
>>>
>>> On 5/18/2016 6:38 AM, William P.N. Smith wrote:
>>> Without a good DVM to measure the exact voltage it's hard to be sure, but are you expecting better than 1% accuracy from a cheap UPS?
>> If they are reporting to the precision of a tenth of a degree, it is reasonable to expect at least that precision.
>> --
>> Bill Drescher
>> william {at} TechServSys {dot} com
>> ------------------------------------------------------------------------------
>> Mobile security can be enabling, not merely restricting. Employees who
>> bring their own devices (BYOD) to work are irked by the imposition of MDM
>> restrictions. Mobile Device Manager Plus allows you to control only the
>> apps on BYO-devices by containerizing them, leaving personal data untouched!
>> https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
>> _______________________________________________
>> Apcupsd-users mailing list
>> [hidden email]
>> https://lists.sourceforge.net/lists/listinfo/apcupsd-users
>
> ------------------------------------------------------------------------------
> Mobile security can be enabling, not merely restricting. Employees who
> bring their own devices (BYOD) to work are irked by the imposition of MDM
> restrictions. Mobile Device Manager Plus allows you to control only the
> apps on BYO-devices by containerizing them, leaving personal data untouched!
> https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
> _______________________________________________
> Apcupsd-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/apcupsd-users

------------------------------------------------------------------------------
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
_______________________________________________
Apcupsd-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/apcupsd-users
Reply | Threaded
Open this post in threaded view
|

Re: Accuracy of utility voltage reported by UPS

jdow
In reply to this post by William P.N. Smith
One might note, quietly of course, that hardly any digits are lost in the
translation from volts to volts. And at a tenth volt accuracy you cannot really
verify it to that tenth volt unless you open up the power supply and measure the
voltage at the same place the UPS measures the voltage. A tenth of a volt drop
in wire and plug/socket contacts is not at all unusual. Experience suggests you
might worry more about the battery float voltage. Set too low you do not get a
good charge. Set too high you destroy batteries prematurely.

I'll also note that 3.5 digit (0..2 for the extra half there) multimeters are
pretty cheap. And 4.5 digit multimeters are used by the electric company when
you complain about low voltage. (Strangely my meter and their meter agreed
remarkably well. Flukes are good. {^_-})

{o.o}

On 2016-05-18 06:08, William P.N. Smith wrote:

> While you and I might understand repeatability, accuracy, and precision, without checking the specs I wouldn't expect too much from a cheap UPS.
>
> I'd start with taking repeated measurements from all three boxes and analyze the numbers.
>
> I've seen too many instances where 2 degrees C is reported as 35.6F (where did all those additional significant figures and extra accuracy come from?) to believe that a $60 UPS can read to 1/10 of a volt.
>
> William P.N. Smith
> ComputerSmiths Consulting, Inc.
>
>> On May 18, 2016, at 7:19 AM, bill <[hidden email]> wrote:
>>
>>> On 5/18/2016 6:38 AM, William P.N. Smith wrote:
>>> Without a good DVM to measure the exact voltage it's hard to be sure, but are you expecting better than 1% accuracy from a cheap UPS?
>> If they are reporting to the precision of a tenth of a degree, it is reasonable to expect at least that precision.
>> --
>> Bill Drescher
>> william {at} TechServSys {dot} com
>> ------------------------------------------------------------------------------
>> Mobile security can be enabling, not merely restricting. Employees who
>> bring their own devices (BYOD) to work are irked by the imposition of MDM
>> restrictions. Mobile Device Manager Plus allows you to control only the
>> apps on BYO-devices by containerizing them, leaving personal data untouched!
>> https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
>> _______________________________________________
>> Apcupsd-users mailing list
>> [hidden email]
>> https://lists.sourceforge.net/lists/listinfo/apcupsd-users
>
> ------------------------------------------------------------------------------
> Mobile security can be enabling, not merely restricting. Employees who
> bring their own devices (BYOD) to work are irked by the imposition of MDM
> restrictions. Mobile Device Manager Plus allows you to control only the
> apps on BYO-devices by containerizing them, leaving personal data untouched!
> https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
> _______________________________________________
> Apcupsd-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/apcupsd-users
>

------------------------------------------------------------------------------
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
_______________________________________________
Apcupsd-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/apcupsd-users
Reply | Threaded
Open this post in threaded view
|

Re: Accuracy of utility voltage reported by UPS

William P.N. Smith
Right, but it's counts in some ADC register to displayed AC volts, and some software engineer asked the hardware guy how many volts per count, did the math, and rounded off to 1/10 volt, which bears no relation to the accuracy or precision of the underlying hardware.

Strangely 😜 APC doesn't specify this parameter

William P.N. Smith
ComputerSmiths Consulting, Inc.

> On May 18, 2016, at 3:00 PM, jdow <[hidden email]> wrote:
>
> One might note, quietly of course, that hardly any digits are lost in the
> translation from volts to volts. And at a tenth volt accuracy you cannot really
> verify it to that tenth volt unless you open up the power supply and measure the
> voltage at the same place the UPS measures the voltage. A tenth of a volt drop
> in wire and plug/socket contacts is not at all unusual. Experience suggests you
> might worry more about the battery float voltage. Set too low you do not get a
> good charge. Set too high you destroy batteries prematurely.
>
> I'll also note that 3.5 digit (0..2 for the extra half there) multimeters are
> pretty cheap. And 4.5 digit multimeters are used by the electric company when
> you complain about low voltage. (Strangely my meter and their meter agreed
> remarkably well. Flukes are good. {^_-})
>
> {o.o}
>
>> On 2016-05-18 06:08, William P.N. Smith wrote:
>> While you and I might understand repeatability, accuracy, and precision, without checking the specs I wouldn't expect too much from a cheap UPS.
>>
>> I'd start with taking repeated measurements from all three boxes and analyze the numbers.
>>
>> I've seen too many instances where 2 degrees C is reported as 35.6F (where did all those additional significant figures and extra accuracy come from?) to believe that a $60 UPS can read to 1/10 of a volt.
>>
>> William P.N. Smith
>> ComputerSmiths Consulting, Inc.
>>
>>>> On May 18, 2016, at 7:19 AM, bill <[hidden email]> wrote:
>>>>
>>>> On 5/18/2016 6:38 AM, William P.N. Smith wrote:
>>>> Without a good DVM to measure the exact voltage it's hard to be sure, but are you expecting better than 1% accuracy from a cheap UPS?
>>> If they are reporting to the precision of a tenth of a degree, it is reasonable to expect at least that precision.
>>> --
>>> Bill Drescher
>>> william {at} TechServSys {dot} com
>>> ------------------------------------------------------------------------------
>>> Mobile security can be enabling, not merely restricting. Employees who
>>> bring their own devices (BYOD) to work are irked by the imposition of MDM
>>> restrictions. Mobile Device Manager Plus allows you to control only the
>>> apps on BYO-devices by containerizing them, leaving personal data untouched!
>>> https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
>>> _______________________________________________
>>> Apcupsd-users mailing list
>>> [hidden email]
>>> https://lists.sourceforge.net/lists/listinfo/apcupsd-users
>>
>> ------------------------------------------------------------------------------
>> Mobile security can be enabling, not merely restricting. Employees who
>> bring their own devices (BYOD) to work are irked by the imposition of MDM
>> restrictions. Mobile Device Manager Plus allows you to control only the
>> apps on BYO-devices by containerizing them, leaving personal data untouched!
>> https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
>> _______________________________________________
>> Apcupsd-users mailing list
>> [hidden email]
>> https://lists.sourceforge.net/lists/listinfo/apcupsd-users
>
> ------------------------------------------------------------------------------
> Mobile security can be enabling, not merely restricting. Employees who
> bring their own devices (BYOD) to work are irked by the imposition of MDM
> restrictions. Mobile Device Manager Plus allows you to control only the
> apps on BYO-devices by containerizing them, leaving personal data untouched!
> https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
> _______________________________________________
> Apcupsd-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/apcupsd-users

------------------------------------------------------------------------------
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
_______________________________________________
Apcupsd-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/apcupsd-users
Reply | Threaded
Open this post in threaded view
|

Re: Accuracy of utility voltage reported by UPS

Jan Ceuleers
In reply to this post by Jan Ceuleers
On 17/05/16 19:44, Jan Ceuleers wrote:
> I realise that these are consumer-grade UPSes, but are they supposed to
> be this bad at measuring utility voltage?

Thanks to everyone who replied; the answer to the above question seems
to be "yes".

I would add that these UPSes only ever report the utility voltage in 2V
increments. So the next step up from 234.0 VAC is 236.0 VAC, etc.

------------------------------------------------------------------------------
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
_______________________________________________
Apcupsd-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/apcupsd-users