[hpsdr] Fwd: Trimble Thunderbolt & Excalibur

Tue Apr 27 00:29:05 PDT 2010

Graham

I hope you don't mind that I've forwarded this to the reflector. It may help someone else.

73

Chris
G4NUX

Begin forwarded message:

> From: Graham / KE9H <KE9H at austin.rr.com>
> Date: 27 April 2010 04:42:50 GMT+01:00
> To: Chris Smith <chris at vspl.co.uk>
> Subject: Re: [hpsdr] Trimble Thunderbolt & Excalibur
> 
> Chris:
> 
> I am glad you got to the bottom of the issue.
> I have a 100 MHz bandwidth scope (samples at 1 GHz.)
> and I have learned to always look for aliasing problems.
> 
> Enjoy your Excalibur.
> 
> --- Graham
> 
> ==
> 
> Chris Smith wrote:
>> Graham
>> 
>> Thanks for the info re the TXCO.  AA8K (no name) said that his was 10.5Hz out compared to WWV & his Tbolt so I will apply the trim tool and set mine up to beat with my Tbolt.
>> 
>> However, I have egg on my face regarding the 434.8Hz signal I observed. I didn't say that I was using a software 'scope - a Bitscope from 6 years ago - which samples at 40MS/s. This means that the 10MHz signal was being sampled at approx 4 samples / cycle - it seems that this is a bit too close to the Nyquist folding frequency and I was seeing the result of aliasing. I'm theorising that the sampling frequency is not exactly 40MS/s but differs by +/-434.8Hz. Hence the "ghost" frequency?
>> 
>> A very good friend Mick, G4EFO, kindly lent me his steam CRO which immediately put my mind at rest as it didn't show the aliasing artefacts. Phew.
>> 
>> This doesn't explain why the ghost didn't turn up on TP2 but that may have something to do with the greater amplitude of the signal at TP2 and hence a greater number of bits from the ADC in the 'scope? I don't know enough about that side of things. I'll leave that to you geniuses. 
>> So, thanks again for your help. I can sleep easy tonight.
>> 
>> Cheers & 73
>> 
>> Chris
>> G4NUX
>> 
>> 
>> On 26 Apr 2010, at 15:03, Graham / KE9H wrote:
>> 
>>  
>>> Chris:
>>> 
>>> Although the typical unit came from the factory within 1 Hz, I have had reports of
>>> two or three that did come 10 Hz off frequency.  I suggest you set it to agree
>>> with your Thunderbolt, and see if it moves again.  I have not had any reports
>>> of problems after re-setting the Excalibur oscillator.
>>> 
>>> There is no obvious source of 434.8 Hz signal on Excalibur.  Check internal power rails for
>>> oscillating regulators?   5 Volts drives the Oscillator.
>>> 
>>> I suspect a switching power supply somewhere is involved.
>>> Look for ground loops involving the output,  or perhaps the scope, or whatever you are driving
>>> with the output connector.
>>> 
>>> If it is present at TP2, and no ground loop problems with the scope, and clean 5 Volt rail,
>>> then it has to be coming out of the Oscillator.  Perhaps a bad Crystek Oscillator?
>>> 
>>> --- Graham / KE9H
>>> 
>>> ==
>>> 
>>> 
>>> Chris Smith wrote:
>>>    
>>>> ***** High Performance Software Defined Radio Discussion List *****
>>>> 
>>>> Hi
>>>> 
>>>> OK, after consulting with Warren on the time-nuts list and running the Lady Heather software, it seems that my Tbolt is working correctly but the temperature reference is being displayed wrongly.
>>>> 
>>>> I still have a couple of problems:
>>>> 
>>>> 1) The spec for Excalibur says that the on-board TXCO (CXOH20) come from the factory set to within 1Hz of 10MHz. I haven't touched the adjustment on the TXCO but if my Tbolt is working OK then the TXCO on my Excalibur is out by some 10 or so Hz as the phase lock LED is alternating red/green at that sort of rate.
>>>> 
>>>> 2) My 'scope is showing that there is another frequency present at the BNC output of Excalibur. A frequency meter shows this to 434.8Hz at 350mV p-p. This waveform is modulating the 10MHz signal.  This modulating voltage is not present at TP2 but is present at TP3, TP4 & pin 2 of J3. Curious. Could there be a problem with the 74AC04 chip?
>>>> 
>>>> I've checked all the voltage rails in Excalibur, especially the 3v3 rail, and they are within a few mV of nominal and are clean. So this isn't a problem with power rails.
>>>> 
>>>> Anybody got any ideas? It's driving me nuts!
>>>> 
>>>> Cheers & 73
>>>> 
>>>> Chris
>>>> G4NUX
>>>> 
>>>> 
>>>> On 19 Apr 2010, at 14:20, Tom Holmes, N8ZM wrote:
>>>> 
>>>>       
>>>>> HI Chris...
>>>>> 
>>>>> If you are not already familiar with it, there is a list like this one
>>>>> called time-nuts, a bunch of folks who are really into time and frequency
>>>>> measurements. There are several on that list who will be able to tell you
>>>>> more than you'd ever want to know about the T-bolt.
>>>>> 
>>>>> You can subscribe at time-nuts at febo.com.  They generally won't know about
>>>>> HPSDR, but your fundamental questions about the T-bolt will get responses
>>>>> for sure.
>>>>> 
>>>>> Regards,
>>>>> 
>>>>> Tom Holmes, N8ZM
>>>>> Tipp City, OH
>>>>> EM79xx
>>>>> -----Original Message-----
>>>>> From: hpsdr-bounces at lists.openhpsdr.org
>>>>> [mailto:hpsdr-bounces at lists.openhpsdr.org] On Behalf Of Chris Smith
>>>>> Sent: Monday, April 19, 2010 9:01 AM
>>>>> To: HPSDR Reflector
>>>>> Subject: [hpsdr] Trimble Thunderbolt & Excalibur
>>>>> 
>>>>> ***** High Performance Software Defined Radio Discussion List *****
>>>>> 
>>>>> Hi
>>>>> 
>>>>> I wonder if some kind soul can throw some light on a problem I'm having with
>>>>> my Trimble Thunderbolt please?
>>>>> 
>>>>> I bought the unit from an outlet in California last November but, due to
>>>>> circumstances, have only just got around to using it with Excalibur. The first thing I noticed was the Temp pane in TBOLTMON shows -30.8C. Jeff
>>>>> Cook G0AFQ sent me a screen capture of his set-up which shows a temp of
>>>>> 36.7C - more like what I would expect to see. TBOLTMON doesn't show any
>>>>> Critical or Minor alarms and the Lat, Long & Alt are all nominal.
>>>>> 
>>>>> The next thing I checked was the output from the 10MHz osc. The amplitude
>>>>> displays serious jitter of several 10s of mVs.
>>>>> 
>>>>> The PPS output displays a 10uS pulse at 1 second intervals, so that appears
>>>>> to be working.
>>>>> 
>>>>> I connected this signal to Excalibur and jumpered for external osc. This is
>>>>> when I became seriously depressed. The phase lock lights alternate red/green
>>>>> at several 10s of Hz. I can only assume that either I have made a serious
>>>>> error in constructing Excalibur or the Thunderbolt output differs from the
>>>>> TXCO on Excalibur by several 10s of Hz.
>>>>> 
>>>>> I've checked the power rails to the Thunderbolt. All were within a few mV of
>>>>> nominal and clean on a 'scope. So it isn't a problem caused by the power
>>>>> supply.
>>>>> 
>>>>> Has anyone else seen these problems? Can anyone suggest a cure?
>>>>> 
>>>>> I'm happy to approach the supplier of the T/B and see if they will exchange
>>>>> the unit but I thought I'd pick a few brains first.
>>>>> 
>>>>> Cheers
>>>>> 
>>>>> Chris
>>>>> G4NUX              
>> 
>> 
>>  
> 