[ANSS-netops] Fwd: Re: [Managers] Fwd: Fwd: Fwd: bug in Reftek "Model 16X" Garmin GPS receiver (details from Reftek)
Antonio Sanchez
A.Sanchez at reftek.com
Fri Jul 15 22:20:56 UTC 2011
REF TEK would like to add little bit more history on the 1 second GPS
problem using Garmin "Puck" 16x Model
.- Troubleshooting has been performed since it was reported
back in December 2010 by PASSCAL when they were running bench tests.
PASSCAL reported seeing some discrepancies that seemed to
follow GPS Firmware 3.30. REFTEK did not see the same problem with
similar bench tests.
REF TEK requested PASSCAL to send specific DAS and GPS's
showing discrepancies after we could not duplicate the problem.
Further bench tests included downgrading/upgrading Garmin
firmware from 3.0 to 3.6
The possible 1 second discrepancy became obvious and
appeared more often under 3.6 firmware and was immediately reported to
Garmin.
Garmin's initial Response :
"What's happening is that there is a tendency for the latency between
the rising edge of PPS and the onset of the NMEA 0183 record to
gradually increase over time. We identified the cause of this
increasing latency and we have fixed the problem. This was
introduced after version 2.80 when we added the PGRMID sentence.
Handling of this sentence should have been done at power up, but was
done instead every second. To get the information, it involves going
through the internal database. As the database grows in size, the
latency increases. The handling of this ID information is now done only
at power up and so the latency issue is no longer present."
This problem could potentially affect the Garmin 16x GPS but it is not
consistent and doesn't always show up in units running side by side.
The discrepancy messages logged by the DAS indicate that the DAS has
detected an inconsistency in time from the GPS. If these messages show
up four consecutive GPS cycles, the DAS changes its time to match the
GPS. A single or periodic discrepancy message doesn't necessarily
means that the DAS timing is off. So, if that message repeats 4 times
in a row then the 130 will change its time to match the GPS. If the GPS
is on duty cycle mode ( normal ) then this process will take 4 hours.
If the GPS is in continuous mode then it takes about an hour. One way to
detect if your 16X is currently affected is by checking the DAS SOH
files for discrepancy messages
REF TEK strongly recommends upgrading Garmin 16x models to 3.70 firmware
as soon as is convenient. Please contact our support team,
http://support.reftek.com or email at support at reftek.com, for
information on how to upgrade Garmin 16x units or options for REF TEK to
do this for you.
Antonio
Antonio Sanchez
REFTEK Customer Support Manager
Refraction Technology Inc
1600 10th Street Plano TX 75074 USA
Ph 214 440-1265
Ph 214 440-1271 (direct)
http://www.reftek.com <http://www.reftek.com/>
http://support.reftek.com
________________________________
From: anss-netops-bounces at geohazards.usgs.gov
[mailto:anss-netops-bounces at geohazards.usgs.gov] On Behalf Of Bridget
O'Neill
Sent: Friday, July 15, 2011 12:23 PM
To: anss-netops at geohazards.usgs.gov
Subject: [ANSS-netops] Fwd: Re: [Managers] Fwd: Fwd: Fwd: bug in Reftek
"Model 16X" Garmin GPS receiver (details from Reftek)
Hello -
Kent Anderson suggested I forward this thread to this forum.
-------- Original Message --------
Subject:
Re: [Managers] Fwd: Fwd: [ANSS-netops] Fwd: bug in Reftek "Model 16X"
Garmin GPS receiver (details from Reftek)
Date:
Fri, 15 Jul 2011 11:17:39 -0600
From:
Bridget O'Neill <brit at passcal.nmt.edu> <mailto:brit at passcal.nmt.edu>
To:
Bruce Beaudoin <bruce at passcal.nmt.edu> <mailto:bruce at passcal.nmt.edu>
CC:
Robert Busby <busby at iris.edu> <mailto:busby at iris.edu> , Kent Anderson
<kent at iris.edu> <mailto:kent at iris.edu> , croker at usgs.gov, James Gridley
<James.Gridley at iris.edu> <mailto:James.Gridley at iris.edu> ,
managers at passcal.nmt.edu, woodward at iris.edu, dietz at usgs.gov
One other thing I should have added -
In the SOH log messages, if the initial EXTERNAL CLOCK IS UNLOCKED
message is missing all the time and you are NOT using a 16x GPS with 3.7
firmware - that is an indication that the DAS power to the clock is
stuck ON. I believe it is U2 on the RT130 Lid board that needs to be
replaced to fix that. We have a check for this issue in our routine
bench testing of RT130's when they return from the field. For some idea
of the frequency of this issue - we've made about 15 of these repairs so
far this year.
- Brit
On 7/15/11 11:00 AM, Bridget O'Neill wrote:
Hi, Everyone -
I just brought this up at last weeks PASSCAL manager coordination
meeting. There was a problem with the 16x receivers that we noted in
January after updating the firmware on the few that we had at that time.
The DAS logs had numerous possible 1 second discrepancies, i.e:
INTERNAL CLOCK POSSIBLE DISCREPANCY OF -1 SECONDS
We did not let any of our 16x, or 15x go to the field with this
firmware. (We are using 15x receivers from Garmin to repair our older
style Reftek GPS, but they did have this same issue.)
We contacted Reftek. Initially they didn't see the problem in 16x's
that they had on hand. We sent them several of our GPSs and DASs and
then they did confirm the problem. I think they then confirmed the
problem in new 16x's they had, presumably after putting the _then_
latest Garmin firmware on them. They contacted Garmin, and Garmin has
recently released new firmware, version 3.7, with this issue fixed. We
have not seen any further occurrence of this problem in our testing at
PASSCAL.
It is also my understanding, from Reftek, from their conversations
with Garmin, that this bug was introduced after the 2.7 firmware. If
your 16x has the 2.7 firmware it is supposedly ok. Reftek said that
they were shipping 16x's with 2.7 firmware between the discovery of the
issue and Garmin's release of the 3.7 firmware.
If you look on the back of your hockey puck style GPS, you can see
whether it is a 16-HVS or a 16x-HVS. We have not seen any problems with
the non-x 16-HVS.
There is something else to be aware of with the 3.7 firmware though.
The clock locks very fast. Normally, the log messages you would have
seen from the DAS during the clock On cycle were:
EXTERNAL CLOCK POWER IS TURNED ON
EXTERNAL CLOCK WAKEUP
EXTERNAL CLOCK IS UNLOCKED
EXTERNAL CLOCK IS LOCKED
(approximately 20 minutes of phase error messages)
EXTERNAL CLOCK IS UNLOCKED
EXTERNAL CLOCK POWER IS TURNED OFF
The time between the power turn on and the DAS first polling the clock
for its status, and getting the first unlocked message, is about 5
seconds. Now the clock usually (but not 100% of the time) locks within
that five seconds, so you don't get the first unlocked message. This
produces a pattern in the logpeek trace of the clock lock/unlock cycle
that is different from what people are used to seeing. However, I have
confirmed with Reftek that there is no problem there. If the clock is
shielded somewhat, so that it takes longer to lock, then you will get
the old familiar pattern of messages and sawtooth (rather than zigzag)
trace in logpeek.
I hope this helps. Please feel free to contact me if there are
questions, or we can be of any assistance.
- Brit
--
Brit O'Neill
Hardware Maintenance
PASSCAL Instrument Center
100 East Rd.
Socorro, NM 87801
Tel: 575-835-5080
Fax: 575-835-5079
On 7/15/11 9:18 AM, Bruce Beaudoin wrote:
I'm just sending this to managers at this point. Please review and if
this is indeed a problem for us (and I'm suspecting that it is) then can
Brit or MikeF be prepared to discuss at Monday morning meeting.
Thanks, Bruce
-------- Original Message --------
Subject:
Fwd: [ANSS-netops] Fwd: bug in Reftek "Model 16X" Garmin GPS receiver
(details from Reftek)
Date:
Thu, 14 Jul 2011 16:04:11 -0600
From:
Kent Anderson <kent at iris.edu> <mailto:kent at iris.edu>
To:
James Gridley <James.Gridley at iris.edu> <mailto:James.Gridley at iris.edu> ,
Bruce Beaudoin <bruce at passcal.nmt.edu> <mailto:bruce at passcal.nmt.edu>
CC:
Bob Woodward <woodward at iris.edu> <mailto:woodward at iris.edu> , Robert
Busby <busby at iris.edu> <mailto:busby at iris.edu>
Hi Guys,
Not sure if you were aware of this, but looks like there is a problem
with the Garmin GPS (model 16x). I believe these are the units we
bought to overcome the problems with the previous SiRF problems in 08.
Just wanted to make sure you were in the loop on this. I'm sure you
guys are all over this already.
Kent
Begin forwarded message:
From: "David S. Croker" <croker at usgs.gov>
Date: July 14, 2011 3:52:33 PM MDT
To: anss-netops at geohazards.usgs.gov
Subject: [ANSS-netops] Fwd: bug in Reftek "Model 16X" Garmin GPS
receiver (details from Reftek)
All, below is an exchange we had after trying to work out a timing error
with a couple of Jim Cullen's sites. Sorry for the bad news....
Dave
Below are the details about a Reftek timing bug, as related by Ian
Billings at Reftek.
Refteks running with Model 16X GPS may or may not have a 1 second timing
error. Once they DO have the timing error, it will remain 1 second wrong
until the GPS firmware is updated. So even if we figure out which
stations have the bad GPS module, we might not know if/when the
timestamp went bad...
The only way to tell if you have a Model 16X is to look at the hockey
puck.
Lynn
Subject: RE: bug in "Model 16X" GPS receiver (Garmin)? [USGS]
Date: Thu, 14 Jul 2011 14:49:25 -0500
From: "Ian Billings" <I.Billings at reftek.com>
To: "Lynn Dietz" <dietz at usgs.gov>
Hi Lynn,
Yes there is a problem with Garmin modules 16X, their Garmin firmware
between 2.50 & 3.70. Garmin have recently fixed this and we have tested
it and ok'd it. The fixed version is 3.70.
Unfortuantely REF TEK did not track 130-GPS serial number to Garmin
model number, therefore one will need to look at the bottom of the unit
to know if it is a 16X. This was a major error on REF TEK's part as
there is no way now for support to contact users that maybe effected.
The error is an exact 1sec difference. Once a clock gets into the state
of being in error it will not come out of it until the firmware is
replaced. However it can take units months to get to this point which
makes it difficult to determine if current units installed are in error
yet.
The 3.70 firmware is available on the Garmin web site. I have also
attached a diagram of the cable we use to upgrade the units. You will
see that a serial port is needed. I will be in your area on the 28th of
this month so we could discuss this more then and I could also upgrade
any units you have on hand.
Let me know if you have ant additional questions.
Ian Billings.
________________________________
From: Lynn Dietz [ mailto:dietz at usgs.gov ]
Sent: Thu 7/14/2011 1:58 PM
To: Ian Billings
Cc: oppen at usgs.gov; dietz at usgs.gov; croker at usgs.gov
Subject: bug in "Model 16X" GPS receiver (Garmin)?
Hi Ian,
Jim Cullen of PG&E just told me that Reftek has confirmed that
there's a bug in the "Model 16X" GPS unit (black hockey puck from
Garmin). Can you give me any details about the timing errors that
result from this bug?
I have observed a timing difference of approximately 1 sec between
co-located Reftek130 and analog data from the PG&E network, but Jim
says he's observed differences of 0.8 sec. So it appears that the
timing error is not exactly a whole second.
I'm not sure if NCSN is using any "Model 16X" GPS units, but I know
we have some black hockey pucks out there. Is there a way to tell
from SOH logs if a Reftek130 is using a buggy Garmin?
Thanks,
Lynn Dietz
__________________________________________________________
USGS - Earthquake Science Center - NCSN Field Operations Manager
David S. Croker office (650)
329-4697
345 Middlefield Rd, MS 977 fax (650) 329-4732
Menlo Park, CA 94025 cell (650) 465-4334
email: croker at usgs.gov
Quake info: http://earthquake.usgs.gov/
USGS URL: http://www.usgs.gov/
_______________________________________________
Managers mailing list
Managers at passcal.nmt.edu
https://webmail.passcal.nmt.edu/mailman/listinfo/managers
--
Brit O'Neill
PASSCAL Instrument Center
100 East Rd.
Socorro, NM 87801
Tel: 575-835-5080
Fax: 575-835-5079
_______________________________________________
Managers mailing list
Managers at passcal.nmt.edu
https://webmail.passcal.nmt.edu/mailman/listinfo/managers
--
Brit O'Neill
PASSCAL Instrument Center
100 East Rd.
Socorro, NM 87801
Tel: 575-835-5080
Fax: 575-835-5079
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://geohazards.usgs.gov/pipermail/anss-netops/attachments/20110715/d8e8cc8d/attachment-0001.html>
More information about the ANSS-netops
mailing list