This article documents VK1OD's implementation of a Trimble Thunderbolt GPS Disciplined Oscillator.
It is with some trepidation that I embarked on this project, as I have an Exactime GPS time reference that depends on a Trimble GPS which did not roll over properly at the first GPS week epoch, and for which Trimble did not offer a firmware fix. With a track record for not understanding or implementing GPS properly at that time and abandoning customers, was the Thunderbolt a better product?
Fig 1 shows the front view of the cased Thunderbolt and its power supply. The case used was an Altronics H 0476.
Fig 2 shows a view of the rear and internals of the case. The blue dots on the GPS cover holes, reducing air flow which compromises stability. The power supply is internally fused, so there are no fuses, just a DPDT rocker power switch (Altronics S 3217).
The Thunderbolt was purchased with power supply (both used) from eBay seller rf-buy2008. The Thunderbolt had a serious problem on delivery, it would raise an oven alarm which did not go away unless the power was cycled. The seller was almost no help, he was slow to respond to email and gave no material assistance. The existence of the fault is possibly the reason why the unit may have been withdrawn from service. I would not recommend this source.
It turned out that the fault was caused by a bad soldered joint on the Thunderbolt PCB pins for the power connector.
As supplied, there was a gross amount of solder flux around the connector area, excessive solder on several pins (a sure sign of soldering failure), and a couple of pins where the solder had obviously not wetted the pins. Removing as much solder as possible, applying a little liquid flux, then resoldering the pins appears to have solved the problem. These connector pins were probably hand soldered in production, were probably reworked, possibly by the seller.
Fig 2 shows the repaired power supply connector pins.
Fig 4 shows graunch marks around the coax connectors probably made by using the wrong tools to remove the nuts. Again, botched rework, possibly by the seller.
Once repaired, the device seems to work properly.
Fig 5 shows a screen clip from Lady Heather monitoring software. During testing, it became evident that the Thunderbolt is quite sensitive to temperature changes, and especially to draughts. A couple of holes in the case were covered by self adhesive dots and the improvement was noticeable. The sensitivity to draughts guided selection of a case to one with reduced opportunity for draughts.
The system draws 20W from mains when it is cold, and that drops back to about 7W when the over reaches temperature. The case does not need to be ventilated to permit that dissipation without excessive temperature rise, the spot above the power supply has the highest surface temperature and it is 10° above ambient.
I will probably develop an AVR based module to monitor the GPSDO data port, and display one or two LEDs to indicate that the unit is working. I will leave more detailed problem analysis to PC based software.
The power connector parts for the Thunderbolt GPS are available from Element14, 135-5950 and 146-2822.
VK1OD on the 'net. I appreciate feedback, click on the ... in webm...@vk1od.net for my email address.
© Copyright: Owen Duffy 1995, 2012. All rights reserved. Disclaimer.