Trimble Thunderbolt - Cheating? Aging product? Total lack of technical support

[max-bit]

Situation:
Trimble Thunderbolt Receiver ..... (GPSDO 10 MHz)

Symptom:
The date passed by the receiver backed by almost 20 years! Real date 1 August 2017 at the receiver display 15 December 1997....what ?   
OK
Probably a bug in the software, just update the firmware and it should be OK.
But no, here is the information from the support ....

Mr Brad (Tech suport Trimble)
Unfortunately, the unit that you have is experiencing the WNRO issue.  The WNRO issue is not fixable via a fw update.  the unit will need to be replaced. 

Can it be exchanged for another device under technical support?
NO !!!
Answer tech :
"Unfortunately, you will need to purchase a new device"
Sorry
Buy it new
What the Fuck !   
The device is IC flash and you (only) need an easy IC firmware update, but Trimble has it in the ass.

ONLY BUY NEW !!!
==============================================================

Conclusions :
DO NOT BUY GPSDO TRIMBLE DEVICES. NEVER

Include documentation "problems" of trimble devices.

[Hydron]

I have also had personal experience of bugs in their Resolution T timing receivers. Due to Trimble making incorrect assumptions about a RESERVED field in the navigation message, when it was changed by the USNO the receivers started losing lock once every 12.5 minutes.

That one was fixable via firmware flash (took a few months for the fix to be released), but that's cold comfort when you've got them inside non-internet-connected equipment in hundreds of secure sites all over the country....

[max-bit]

In this case, do not issue new firmware

[Hydron]

Losing lock (even briefly) was unfortunately not really acceptable behavior in this case, so the firmware upgrade was the only solution. Thankfully the equipment connected to it could deal with it well enough that it wasn't critically urgent and could be put off until the next time a technician was scheduled to be there, but it was a close one.

[texaspyro]

Week number rollover where the 10 bit week number broadcast by the satellites wraps from week 1023 to week 0 is a very common problem with GPS receivers.  It happens every 19 or so years.  Most receivers have a week number offset compiled into their firmware so that the receiver is good for 19 years until the next rollover + compensation offset occurs.  The next generation GPS satellites have a 13 bit week counter,  but it will probably be 10 years for the current satellite constellation to be replaced and new GPS receivers that know about 13 bit week numbers are available.

Thunderbolts have been an obsolete product  no longer supported by Trimble for many years.  The date rollover issue is documented in their user manuals.  It would be nice of them to offer a firmware update, but I can't really fault them for not offering a firmware fix for such an old product.

Lady Heather has an automatic rollover detection  and compensation feature built in.  After 10-15 seconds of dates before 2016,  it corrects the date forward 1024 weeks.  You can also force the rollover mode on from the command line so you don't have to wait 15 seconds for the date to fix.

Most NTP drivers and GPSD know about rollover issues and have code to work around it.

[K1JOS]

 check website of KO4BB --- http://www.ko4bb.com/getsimple/index.php

He makes a nice inexpensive display for the Thunderbolts and will have corrections built into his firmware

Jerry NY2KW

[branadic]

Bought a Trimble Thunderbolt E last year via a german distributor. Device received a few weeks later.
When monitoring the ouput level of the 10MHz signal I found that output voltagel was way out of spec and 10MHz everything but stable. So I contacted the distributor, but as he was only a distributor he couldn't help and forwarded information to Trimble.
After some mail comversation I sent him the GPSDO with a description of the failure I found.

He sent the GPSDO back to Trimble. I'm not sure if they do really have qualified employees, but they were not able to understand the problem, which I don't understand as they only had to monitor output voltage to see the problem of the unit.
However, weeks later without a feedback from Trimble the distributor decided to order a replacement from Trimble and sent me this unit, which is working now quite well in our lab.
Conclusion: Trimble support service is very annoying and everything but customer friendly.

-branadic-

[max-bit]

Trimble is well aware of the problem and can fix it by updating the firmware.
But better sell a new product for thousands of dollars! Better for the company.
But they forgot the fraudsters that buyers are not losers.
And they are generally well-known electronics technicians.

Branadic
Have you bought a new one? Device, so grace does not do that they have (they must!) You replace the device with new ones.
But as you wrote, TRIMBLE has been causing problems !?
My device is already several years old, and TRIMBLE believes that it can have a customer's on ass.

[WastelandTek]

good to know

[ebastler]

Unfortunately, the unit that you have is experiencing the WNRO issue.  The WNRO issue is not fixable via a fw update.  the unit will need to be replaced. 

Can it be exchanged for another device under technical support?
NO !!!
Answer tech :
"Unfortunately, you will need to purchase a new device"

Hi Maxbit,
I understand that you are frustrated. But I feel that your post is missing some relevant information:

• When was the product manufactured? It is normal for manufacturers to discontinue support of legacy products at some point in time -- provided that the suppport period is reasonably long.
• I assume you are using this unit with some kind of front end. Can't you fix the week roll-over in the front-end, so you can continue to use the Thunderbolt?

[edpalmer42]

I looked for some kind of EOL document for the Thunderbolt, but couldn't find one.  It looks like it ceased to be a 'current product' sometime before 2010.  The typical Thunderbolt sold on ebay and elsewhere - i.e. the plain package - isn't even the retail version of the Thunderbolt with the red label.  Who knows what the actual requirement was for the plain package version?  Maybe they didn't even care about the year.

It doesn't surprise or upset me that it has this kind of problem since it's a GPS problem rather than a Trimble problem.  Every GPS device that tries to extract the date from the satellites will have this type of headache eventually.  As ebastler stated, support eventually ends for every product.

It's also worth noting that this issue doesn't affect the Thunderbolt's performance as a GPSDO, i.e. providing 1 PPS and 10 MHz signals locked to GPS.  It only affects the year.  The time and the rest of the date are still correct.  Lady Heather already compensates for it.  I've seen one report that a Linux fix has been implemented.  I haven't seen anything official from ntp.org.

Ed

[cdev]

One could just stick a newer more modern GPS in there hidden in some corner to provide the NEMA data.

[max-bit]

Thunderbolt sold on eBay and Thunderbolt E (lab kit) are the same products.
Of course, Trimble does not care to support but sell new ones.
Of course, the problem is with the GPS (specification) but it can solve the software update.
However, Trimble does not care about that.

And here is the main problem.

[max-bit]

To make it more ridiculous, after asking a technical question whether there is a possibility to reprogram the flash memory itself (, contact stopped.
Here is TRIMBLE suport
The device is an am29F400bb flash memory. So reprogramming memory is no problem.

[ebastler]

To make it more ridiculous, after asking a technical question whether there is a possibility to reprogram the flash memory itself (, contact stopped.
Here is TRIMBLE suport
The device is an am29F400bb flash memory. So reprogramming memory is no problem.

Yes, Trimble sucks, in your opinion. We got that now. What else do you expect from this thread?

[max-bit]

Firmware to update.

[branadic]

Branadic
Have you bought a new one? Device, so grace does not do that they have (they must!) You replace the device with new ones.
But as you wrote, TRIMBLE has been causing problems !?
My device is already several years old, and TRIMBLE believes that it can have a customer's on ass.

Yes, we bought a new unit via distributor in germany from Trimble. And yes, the defect unit was replaced by a new unit.
Trimble was not able to understand the problem I discribed (output voltage level is not within spec), instead they told us that the unit is working fine. My guess is, they haven't even touched nor tested the defect unit that was sent back to them.
However, the replacement is now working fine and within spec.

-branadic-

[ebastler]

What else do you expect from this thread?

Firmware to update.

You expect the audience on this thread to provide you with firmware?!   

Come on, man. You bought a third-hand product cheap on ebay. You claim that it is "the same as the Thunderbolt E", but it is probably an earlier version (Thunderbolt or Thunderbolt II?), which Trimble no longer supports with updates. You never had a business relationship with Trimble, it appears. So either you solve this problem yourself, in whatever host you use the Thunderbolt with, or you buy a new unit, Trimble or not. Simple as that.

[texaspyro]

Thunderbolt sold on eBay and Thunderbolt E (lab kit) are the same products.

No they are not!   Different design.  Different receiver.  Different firmware.

[WastelandTek]

You expect the audience on this thread to provide you with firmware?!   

get coding you slackers!

[cdev]

How long should they fix things like that?

If its easy I think they should for at least 20 years. Or make the software modular or user-configurable.

[texaspyro]

How long should they fix things like that?

If its easy I think they should for at least 20 years. Or make the software modular or user-configurable.

The Thunderbolt design is pretty close to 20 years old...

[bingo600]

@max-bit i'm quite sure you can get the tbolt sold to a user here , for a decent price. If you don't want it anymore.

I have 3 tbolt's , and i got them for 10MHz & 1PPS outputs.

The Week Rollover doesn't affect the frequency outputs , just the date sent.

I'm using Lady Heather, to dicipline my tbolt's , and "TP, the author"    even made it automatically detect the rollover. AND show the "corrected date".
See the yellow "ro" after the date , indicating it's rollover corrected.




I'm using the linux version, that runs native X.

/Bingo

I wonder what happens at the next leap second though ... Well see.
And i'm quite sure that it will still output a nice 1PPS & 10Mhz , maybe with a glich around the leap

[metrologist]

I see my Lassen LP's are in the list, but to call for replacements. 

Maybe for the original application (TBolts) it does not matter?

[max-bit]

No replacements only buy new full price maybe ...