[time-nuts] Trimble Thunderbolt GPSDO Troubleshooting

Ryan Stasel rstasel at uoregon.edu
Sun Mar 27 19:59:45 EDT 2016


All,

So, ended up pulling the 232 chip, and it seems to test okay as far as resistance between pins, and the board still only shows 65ohms or so between +5v and ground around that chips pads (support caps, pads, etc). I’m not sure if that’s “normal”, but it doesn’t seem like it is to me. Nothing seems to be getting abnormally warm. =/ I’m half curious if my meter (Keithley 196, and a Fluke 189) is turning on a transistor partially. =/

I do see what appears to be TTL level serial communication on the inputs to that chip. The replacement should be here on Monday, but at this point I'm wondering if my meter might be turning on some transistor (it shouldn't), or...? I would think a shorted cap would itself heat up.

There's a thread online relating to MAX232 type chips suffering from latchup under certain conditions (https://e2e.ti.com/support/interface/etc_interface/f/392/t/233847). But the low resistance between power and ground rails has me a bit concerned. :/

I did leave the unit running for about an hour, and the 10Mhz never came down from its 26hz high. Seeing that, I looked at the adjust pin on the OCXO, and it’s pegged at -10.25V. That also appears on all the outputs from the LT1014. =/ Also in that area is the 5V reference, and checking the output of that shows -0.6V (should be 5V). I’m… really starting to worry this whole thing is toast. All the rails appear fine, so I can’t help but think the previous owner hooked up the power wrong, and either -12V or +12V got applied to the 5V parts. =/ I’m fine replacing these two other LT parts, I’m just starting to worry that something not as easy to replace could be damaged (the FPGA, the Trimble chip, etc). Where is the actual DAC (is that in the FPGA)? Though I suppose if the 5V reference is dead, that would probably prevent the DAC from functioning. I guess the question would be if something is pulling the ref down (below ground), or is it just borked?

Anyway, hopefully I’ll at least be able to hook up serial to it tomorrow evening and see what Tboltmon says.

Thanks again for the help. Let me know if you have any other thoughts for things I can check.

Ryan Stasel
IT Operations Manager, SOJC
University of Oregon

Sent from my iPhone

On Mar 27, 2016, at 8:53 AM, Robert LaJeunesse <lajeunesse at mail.com<mailto:lajeunesse at mail.com>> wrote:

Ryan,

The cap across 1-3 is the one that boosts +5V to +10V, with pin 2 being the +10V result. Thus the pins on this "flying" cap swing 5V p-p (pin 3 is 0 to 5, pin 1 is 5 to 10). The cap across 4-5 is the one that inverts +10V to -10V, with pin 6 being the -10V result. Pin 6 should NOT be at ground. The pins on this second flying cap swing 10V p-p (pin 4 is 10 to 0, pin 5 is 0 to -10).

Inputs to the transmitter stages (10 & 11) should be at valid logic levels, for the ICL232 these are standard TTL levels. Be aware that other part types that do this same function may not have the same logic levels (Vih and Vil) so be careful when substituting.

I did not mean to suggest lifting pins 10 or 11, although you might consider that last, but pins 7 and 14 should idle near -10V since the transmitter inputs have pullups and the drivers invert. What I was suggesting is to make sure pins 8 & 13 were disconnected [just pull off any RS-232 cable] and look for pins 9 and 12 to be at a good high level, say above 4V or so.

FWIW it's just as difficult to get -12V from +12V via an inverter chip as it is to simply use the ICL232 (or MAX232 or similar) part. Add to that the fact that older high-voltage RS-232 parts tend to be power hungry, and require separate packages for transmitter and receiver. Why use 3 parts when 1 works?

Good luck with scoping out the real problem.

Bob LaJeunesse

Sent: Friday, March 25, 2016 at 5:33 PM
From: "Ryan Stasel" <rstasel at uoregon.edu<mailto:rstasel at uoregon.edu>>
To: "Discussion of precise time and frequency measurement" <time-nuts at febo.com<mailto:time-nuts at febo.com>>
Subject: Re: [time-nuts] Trimble Thunderbolt GPSDO Troubleshooting

Hi Bob,

I think I’m going to take a look, since I couldn’t find a spare, so I’ve got a couple on order (and a few days to wait).

So, looking at the datasheet (http://www.intersil.com/content/dam/Intersil/documents/icl2/icl232.pdf), you’re saying check the two caps across pins 1 and 3, and 4 and 5, and make sure they’re swinging. And also check to see if Pin 6 is at “ground” (basically). +V should be at around 10V? Likewise, I’d assume the input to that stage (wherever it comes from) should be a nice logic level input.

And then lastly, you’re saying lift/cut pins 10 (and/or 11, depending), and then look at 7 or 14 and see if they’re sitting at 5V (they wouldn’t sit at +10V)? It’s actually kind of funny that they used this chip, since it’s whole point is to eliminate the need for ±12V… except this board specifically has ±12V! =P

Anyway, I’ll check all this as best I can, or just get frustrated lifting pins, and remove the whole darn thing (like Clint suggested). =P

Thanks all!

-Ryan Stasel

On Mar 25, 2016, at 12:52 , Robert LaJeunesse <lajeunesse at mail.com<mailto:lajeunesse at mail.com>> wrote:

Personally I'd take a few minutes to look closer at the '232 chip with the scope. Are the charge pump caps swinging 5V p-p and 10V p-p? if not swinging at all, trash the chip. Is -V pin voltage about equal to the negative of the +V pin? If so, that's OK. If not might have an output shorted, or a bad chip. Is the +V pin about +10V?  If so, that's OK. If not might have an output shorted, or a bad chip. With no RS232 driving source connected does the receiver output look like a valid 5V logic high level? If not, look for a short on this line. If all these pass and the chip gets hot, replace the chip.

Bob LaJeunesse
_______________________________________________
time-nuts mailing list -- time-nuts at febo.com<mailto:time-nuts at febo.com>
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

_______________________________________________
time-nuts mailing list -- time-nuts at febo.com<mailto:time-nuts at febo.com>
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

_______________________________________________
time-nuts mailing list -- time-nuts at febo.com<mailto:time-nuts at febo.com>
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.



More information about the time-nuts mailing list