Bug 15693 - Spurious AC voltage readings
: Spurious AC voltage readings
Status: RESOLVED FIXED
Product: SB Transporter
Classification: Unclassified
Component: Hardware
: 80
: All All
: P1 normal with 8 votes (vote)
: 8.0.0
Assigned To: Chris Owens
:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2010-02-12 15:07 UTC by Chris Owens
Modified: 2010-08-08 06:54 UTC (History)
4 users (show)

See Also:
Category: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Chris Owens 2010-02-12 15:07:40 UTC
I got the following email from Sean:

People are still seeing those spurious overvoltage errors on transporter. It is understandably very disconcerting to them and they will think there must be something wrong either with their AC or with the TP hardware.

The error is caused by spurious readings from an internal ADC which measures one of the internal DC power rails. I was never able to figure out exactly why it gets these spurious readings but I think it has something to do with how the ADC's control pins are multiplexed (shared) with some other on-board peripheral.

I would suggest removing this failsafe feature entirely. I originally put it in there as a debugging feature, like an assert()... not because I thought that sustained overvoltages would actually be present on people's power supplies. In reality the feature does not provide any protection from common power surges.
Comment 1 Chris Owens 2010-05-17 09:12:12 UTC
Felix notes that he is worried about changing this part of the code.  He would like a reproducible case to test with.
Comment 2 michael123 2010-05-17 11:15:49 UTC
it happens in 100% of the cases in my system when I switch tracks.
Mostly, when I switch more than one, fast.
Also, I think it happens more with high-res (heavy) tracks
Comment 3 Chris Owens 2010-05-24 09:03:13 UTC
Michael, where in the world are you and what voltage do you have there?  Thanks!
Comment 4 michael123 2010-05-24 10:02:09 UTC
(In reply to comment #3)
> Michael, where in the world are you and what voltage do you have there? 
> Thanks!

I am in Israel, this is 230V.
I have Furman SPR16Ei power stabilizer attached, which shows also 229-231 reading all the time..
Comment 5 michael123 2010-05-24 10:45:22 UTC
(In reply to comment #4)
> (In reply to comment #3)
> > Michael, where in the world are you and what voltage do you have there? 
> > Thanks!
> 
> I am in Israel, this is 230V.
> I have Furman SPR16Ei power stabilizer attached, which shows also 229-231
> reading all the time..

BTW, Voltage via the web page is far from being precise:

   Player Model: Transporter
   Firmware: 80
   Player IP Address: 192.168.3.152
   Player MAC Address: 00:04:20:10:0a:34
   Voltage: 226

while I see 229 on Furman.
Comment 6 Felix Mueller 2010-05-28 01:21:41 UTC
From Sean's comment about the spurious readings and the multiplexing I assumed to see values not from the ADC (i.e. in the normal voltage range) but from one of the other multiplexed chips and expected the values to be way above normal voltage. I also would have expected that the spurious readings from that other chip would then happen regardless of TR being in a 110V or 220V area.

But it seems that it only happens when TR is running from 220V. Is that actually true?

On my TR (running a debug firmware and with 220V) I occasionally see values slightly too high. I.e. 272V, 266V, 269V, 276V. The AC over voltage kicks in above 265V. This sometimes happens when TR is just idling and sometimes on a track change.

I'd like to do some more debugging to get a better understanding what is going on and to confirm my findings.

BTW: The voltage reading was never meant to be precise, it is only used to adapt the wiring to the transformers for 110V and 220V. So a reading with an error about +/- 10% is to be expected.
Comment 7 michael123 2010-05-28 02:02:49 UTC
(In reply to comment #6)
> From Sean's comment about the spurious readings and the multiplexing I assumed
> to see values not from the ADC (i.e. in the normal voltage range) but from one
> of the other multiplexed chips and expected the values to be way above normal
> voltage. I also would have expected that the spurious readings from that other
> chip would then happen regardless of TR being in a 110V or 220V area.
> 
> But it seems that it only happens when TR is running from 220V. Is that
> actually true?
> 
> On my TR (running a debug firmware and with 220V) I occasionally see values
> slightly too high. I.e. 272V, 266V, 269V, 276V. The AC over voltage kicks in
> above 265V. This sometimes happens when TR is just idling and sometimes on a
> track change.
> 
> I'd like to do some more debugging to get a better understanding what is going
> on and to confirm my findings.
> 
> BTW: The voltage reading was never meant to be precise, it is only used to
> adapt the wiring to the transformers for 110V and 220V. So a reading with an
> error about +/- 10% is to be expected.

Nice, I am glad you could reproduce the issue.
Reading the voltage reading, it is a minor one..
Comment 8 Julius Dauz 2010-05-28 09:12:08 UTC
According to Peter DeBrass in EMEA, this doesn't really come up. He stated that over the course of time they've only had about 2-3 cases related to this issue.

This information was pooled from the tier 2 agents from our support partners and Bill Linke.
Comment 9 michael123 2010-05-28 09:17:32 UTC
(In reply to comment #8)
> According to Peter DeBrass in EMEA, this doesn't really come up. He stated that
> over the course of time they've only had about 2-3 cases related to this issue.
> 
> This information was pooled from the tier 2 agents from our support partners
> and Bill Linke.

Well,
I have this 'feature' couple of times per month.
And this indeed, happens mostly when I change the tracks (but not always).
Comment 10 SVN Bot 2010-06-02 09:45:02 UTC
 == Auto-comment from SVN commit #6759 to the player repo by fmueller ==
 == http://svn.slimdevices.com/player?view=revision&revision=6759 ==

Bug: 15693
Description: Fix for spurious AC voltage readings
Comment 11 Felix Mueller 2010-06-02 09:47:34 UTC
Update hours worked
Comment 12 michael123 2010-08-08 06:54:54 UTC
Hi
Could you please tell if this fix will be available in next 7.6 release?
Is it possible to grab a beta of (Transporter) firmware?

thanks,