EEVblog Electronics Community Forum
Electronics => Metrology => Topic started by: mcinque on January 12, 2018, 05:47:47 pm
-
Just a quick look inside the nice and useful reference sold by leobodnar.com.
Just tried, very useful to check cheaply my frequency counter calibration.
Installs quickly, finds the GPS fix very fast and uses an essential and robust software utility to configure it.
http://www.microchip.com/wwwproducts/en/PIC18F24K50 (http://www.microchip.com/wwwproducts/en/PIC18F24K50)
https://www.silabs.com/documents/public/data-sheets/Si5328.pdf (https://www.silabs.com/documents/public/data-sheets/Si5328.pdf)
https://www.u-blox.com/en/product/max-m8-series (https://www.u-blox.com/en/product/max-m8-series)
-
See, a MEMS device on PCB layer :)
-
Hello,
I planned to purchase also the mini precision GPS clock from Leo Bodnard
because it allow programming output frequency in broad range and
it have much lower phase noise than lower cost GPS clock from BG7TBL.
Anyway, before to purchase, I asked some info to Leo Bodnard support email,
but no answer after a week or so. I hope he's fine...
So I post my questions here, maybe some owners would have the answers :
1/ Does it support Galileo GPS lock ?
2/ Does the provided control software is Linux compatible ?
3/ Is there a manual available with all specs (a real datasheet with measurements) ?
Thank you in advance
Frex
-
Hi Frex,
1/ Does it support Galileo GPS lock ?
Yes, it supports GPS, Glonass, Beidou and Galileo but I tend to keep Galileo turned off. It degrades high precision timing performance but you need a high quality stable source to see that.
2/ Does the provided control software is Linux compatible ?
You can get Linux version from https://github.com/simontheu/lb-gps-linux (https://github.com/simontheu/lb-gps-linux)
3/ Is there a manual available with all specs (a real datasheet with measurements) ?
All the information is on the website http://www.leobodnar.com/shop/index.php?main_page=product_info&cPath=107&products_id=301 (http://www.leobodnar.com/shop/index.php?main_page=product_info&cPath=107&products_id=301) There is not really much to it.
Cheers
Leo
P.S. I can't see your email, please use support@leobodnar.com for contacts
-
Hello,
Thank you for this quick answer.
As I seen, for Linux it's not a GUI but only by commands line instructions(?).
So now, I just need to make the order ;o)
Note that my two email (01/09 and 02/09) has been sent you
to the right email address. Maybe a spam filter ? can you check ?
(I don't have issue with my email usually)
Regards.
Frex
-
Hi,
A short question to Leo Bodnar:
I have the two channel version of the Precision GPS Reference Clock and I am very happy with it. But why is there no satellite reciption function or acess to the GPS funcions like in the mini Precision GPS Reference Clock? What is the reason for this limitation?
-
Leo, I recently got the MiniGPS, and really like it. I had been using a very old Trimble Thunderbolt, and the Mini is giving me much less trouble. Thanks!
A question for you: I notice what I presume to be a thermal isolation cutout around the Mini's TCXO. How much does that help, compared to leaving the PCB intact?
-
Hi,
can this unit communicate with the u-center software trough the USB port?
Chris
-
I have the two channel version of the Precision GPS Reference Clock and I am very happy with it. But why is there no satellite reception function or access to the GPS functions like in the mini Precision GPS Reference Clock? What is the reason for this limitation?
We will have a unified control application for both clocks at some point and you will get the access to this info/functions too.
can this unit communicate with the u-center software trough the USB port?
Not at the moment, no.
A question for you: I notice what I presume to be a thermal isolation cutout around the Mini's TCXO. How much does that help, compared to leaving the PCB intact?
It reduces the rate of frequency change making it easier for the DPLL BW to follow the drift.
Leo
-
Leo, my dual output unit takes around 5 minutes to obtain lock if it was been unpowered for more than about 1 hour. It is the same even if I mount the patch antenna on my car roof with a completely clear sky view.
Should I replace the capacitor?
Dave
-
Dave,
Supercap should provide enough power to keep timing for 4-5 hours.
Leo
-
Hello,
Just to say thank you to Leo.
I received the mini precision GPS clock and it work great.
The ability to set output to any value from 400Hz to 800M and the low phase noise
output has decided me to choose it instead of Chinese GPSDO (BG7TBL).
A little bit more expensive, but it worth the money !
Regards.
Frex
Note : Anyway, If you serch about what could be improved,
I must say that the ability to switch output to sine instead of square would be great.
Or better, a second dedicated outputs. ;-)
-
There are several docs on internet to calculate the output frequency 1 and 2 , but i wanted a simple formule spreadsheet.
I have the GPSDO with two outputs.
So i came to this simple sheet:
For calculation of the two output frequency's, you need just two frormules:
Output1: formule = ( GPSref x N2_HS x N2_LS ) / ( N31 x N1_HS x NC1_LS )
Output2: formule = ( GPSref x N2_HS x N2_LS ) / ( N31 x N1_HS x NC2_LS )
(SI5328, CKIN1 = GPSref out from the MAX 8 module)
In this sheet a sample of 10 Mhz on output 1 and 1 Mhz on output 2 that works,
and the formules.
The GPS module can deliver an input frequentie up to 10 Mhz to the Si5328, and
is then divided bij N31 for reference to the DSPLL (F3).
*Note, the Fosc frequency has to be between 5 and 6 Ghz for a PLL lock.
I think it is a kind of 5 Ghz wifi VCO ? Can not read any signs on it.
My unit has the following limits for the VCO frequency, to lock the PLL: 4.17 Ghz and 6.68 Ghz.
So the safe limits are between 5-6 Ghz.
-
Needed something for a long time and wasn't sure about the Chinese units, so just ordered one of Leo's. Time to get those counters tuned up!
-
Hello all have both Leo's GPS and the 40 pS pulser, also the SDR radio SDRPlay.
All excellent designs and good SW and support.
HIGHLY RECOMMENDED
Suggest to avoid the Chinese junk knockoffs with zero support.
Bon journee,
Jon
-
Very impressive. The mini arrived in the US two days from shipment in the UK. The good news is I had a chance to verify it was working great. When I beat my counters against WWV I can get about 10E-7 or a bit better on a good day. This is way better. The bad news is I then had to pack it back up so my wife could put it under the tree for Christmas. I'll report back in January.
-
In the meantime, I don't really know how to read the Allen deviation chart; does it mean I've achieved maximum accuracy after about 2000 seconds of operation? How long should one wait to get best performance?
Also, are the antennas magnetic because you get better sensitivity on a metal ground plane?
Thanks!
-
There is lot of information on the net about Allen deviation charts,
But on the short, on such chart can see how a oscilator performs for frequency stability on the short and on the long run.
GPS is on the short time very bad, due to radio path variations and reflections and number of satelites in view.
That can be as bad as 10E-6. But on the long term GPS is very very good, because it follows the ground stations corrections.
In an allen dev chart you can see this performance over time.
A temp controlled oscilator is very good on the short term, but less on the long run.
So if you combine these two, GPS and a TCO you get the best of these.
I have two lucent rubidium oscilators, and a HP5370B counter which i can use to compare with the Bodnar GPSDO.
My GPSDO locks in about 2 to 3 seconds, and is then better than 10E-9
After 10 minutes i get about < 3 E-10 measured with the HP5370, interval ~1 second. ( see graph of the next 10 minutes = 400 samples)
The rubidium clocks are about <5 E-11
And about the magnetic feet , thats is only for easy use on metal, like cars
-
Thanks! That's good to know. From reading about various disciplined oscillators, I was under the impression it could take far longer to get decent accuracy. I keep reading about Allen deviation but it doesn't seem to sink in. Too much math and too little overview. Or, I'm just a dummy.
-
I made a comparison today of my three GPSDO’s: the Bodnar Precision unit, a BG7TBL version, plus a Trimble GPSDO with three both sine and square wave outputs. I used my Anritsu MT8222A analyzer for SA measurements. It has a built-in GPSDO.
All three had clean 10 MHz outputs. The Bodnar unit was nice in being able to provide 450 Hz to 808 MHz outputs. However, some frequencies are not as good of quality. I provide views of 43.2 and 330 MHz as two examples. The BG7TBL FA-2 is the best thing that I have for precise frequency measurement. Using the BG7TBL GPSDO as its reference, the Bodnar unit was measured 164.2 uHz high. The accuracy of that measurement obviously is dependent on both of those devices. The MT8222A measured the fifth harmonic of 190 MHz from the Bodnar unit at 950 MHz as being 0.15 Hz low.
I also tested my BG7TBL 10 MHz DA dated 2019-09-13 on its front panel. It has eight outputs and did not change the quality of GPSDO's feeding it.
The Trimble GPSDO had poor frequency accuracy: 0.425575 Hz high. I include pictures of its circuit board in case anybody knows how to improve that.
-
How do you know it's the Trimble that is off , and not the BG7 ?
I would prob. initially have trusted the Trimble , due to it being a Trimble , and not a non "Home made board"
Edit: The board w the 8-chan is the BG7 (Based on a Trimble Top board) ?
Edit2: For how long has the boards been powered up , an OCXO can take several weeks to settle.
/Bingo
-
The Bodnar unit was nice in being able to provide 450 Hz to 808 MHz outputs. However, some frequencies are not as good of quality. I provide views of 43.2 and 330 MHz as two examples.
These look very ugly. How exactly are you measuring this? I.e. what instruments, refernces, settings and measurement procedure?
I have quickly thrown a mini GPS clock set by config software to default 330MHz on a E4406A, set to 40Hz RBW - using its mediocre internal oscillator (all my Rubidium and OCXO references are at work.) I have used standard magnetic puck antenna right on top of E4406A, inside my house.
Here is what I have after 100 averages (took about 10 minutes)
Phase-noise wise, what you see here is the noise of E4406A, mini GPS clock is cleaner - you should see about -70dBc/Hz at 10Hz offset off 330MHz carrier. Carrier offset is not precisely 330.0000000MHz since E4406A is not locked to anything and has not been calibrated for ages.
Cheers
Leo
Edit: I have also added 43.2MHz take. The shape of the carrier on your screenshots look like Blackman-Harris window, so I have used one too.
-
How do you know it's the Trimble that is off , and not the BG7 ?
It is the Trimble vs. BG7TBL, Bodnar, MT8222A, and many other local RF sources which I know are GPS locked. Attached are pictures of the the Trimble's front and rear panels. Planning to get back into it, I left its mounting screws off.
The Trimble was only on for several hours before the measurement, as were the other GPSDO's. They had much better accuracy, agreeing the other known accurate frequency sources. With my common use being at random locations, I need reasonable accuracy after hours of warm-up, not days. The others can achieve that.
-
I made a screenshot of the 10 Mhz and 330 Mhz signal of the Bodnar GPSDO, plotted on a SDR,
not to overload the SDR there is a 40 dB att inserted. Still have a 100 dB range.
it looks normal for me not bad at all.
-
How do you know it's the Trimble that is off , and not the BG7 ?
With my common use being at random locations, I need reasonable accuracy after hours of warm-up, not days. The others can achieve that.
While that says something about your usecase, and preferred unit.
How does it prove that the Trimble is the one that is "Off" ?
Seems to me that a Rubidium is what you might want , since you can't wait for the OCXO to stabilize.
/Bingo
-
How exactly are you measuring this? I.e. what instruments, refernces, settings and measurement procedure?
As previously said, I was using an Anritsu MT8222A with GPS and several other options. See:
https://www.aaatesters.com/pub/media/datasheets/anritsu_bts_master_mt8222a_specifications_spec_sheet_6c115.pdf (https://www.aaatesters.com/pub/media/datasheets/anritsu_bts_master_mt8222a_specifications_spec_sheet_6c115.pdf)
Its settings were shown on the diagrams. For the collection of three, other than the frequency, the settings were the same for all three.
The problem seems similar to those reported on SV1AFN's GPSDO. It only occurs on particular frequencies. I wanted 990 MHz, so I chose to use the third harmonic of a 330 MHz square wave. Attached is what that harmonic looks like. 330 MHz still looks like what was previously sent. Padding its level down to leave additional headroom did not change the display. It just removed the dynamic range shown.
Also attached is what the third harmonic of 331 MHz looks like from the Bodnar unit--not bad.
-
Hallo OCW,
Can you try these settings to get 330 Mhz: maybe you get different result ??
Your settings are to nice, maybe they interfere.
Ref= 4687500
N31= 5
N2H= 11
N2L= 512
N1H= 4
NC1= 4
NC2= dont care
See attach settings that i used:
-
Thanks Wim13,
Those settings produced a much better looking waveform--see the attachment.
I used the settings produced after pressing "Find." It just points to checking the output when going to a new frequency if a clean waveform is needed.
-
Tested the settings you used before, and get indeed lot more spurious, zie attachment.
Some kind of 8.6 Hz fm intefere on something...
So it is usefull to try other combinations if you have some suprious.
Calculations are not difficult but take some time.
Example: Fosc of 5.280 Mhz is in factors, =( 2^11 and 5^7 and 3 and 11 ),
then you see something with 33 or 330 is possible, but 7, 9, 13 etc are not possible in this example.
You have to find then an other Fosc between 5 and 6 Ghz.
The lowest value for N1_HS is 4 and max 11 so for the first devider in this example, the freq. are between 480 Mhz and 1320 Mhz.
The next divider example: has to be an even value, possible output is then 480 - 240 - 120 - 80- 60 - 48 etc.
So that is why i use my simple spreadsheet, then i can easely test all values,
and to avoid frequencies that can interfere with each other, and too rounded values that are related.
-
Wim13, ocw,
Thanks, that's it. Each output frequency has several thousand combinations that result in the same nominal output. But they all have different intermodulation spurs content. Sometimes software picks up not the best combination. I will have a look at this. Interestingly, mini GPS software picks up a different set which is cleaner. My tests were on mini GPS.
You can also try these settings. They also seem to be clean.
If you are looking for the most stable freqeuncy, you will find that GPS clock (and mini GPS clock) converge to its best accuracy after 10-15 minutes. This is mostly due to GPS needing some time to derive its stationary position to better accuracy. Also if you are mobile - don't move the antenna, it will produce phase shift and will require more time to settle again.
Cheers
Leo
-
A look into the power handeling of the two output GPSDO....because, some times no lights ??
Then I noticed that the two leds go off when the PC has been shut down, while there is still an extern power supply connected.
The same as i leave the USB connector inserted and the PC is off, no leds on.
Nothing seems to happen then, with external power switching on.
So i toke a look inside if something was wrong with the power.
The power input has a 5 Volt regulator the LM2940C, see attached datasheet.
This regulator needs min 6 Volt and max 26 Volt. Has a low drop of 0,5 Volt.
The GPSDO should work also on 5 Volts...?
Well on the next level there are two 3,3 Volt regulators, the LLVB (LP5907) 250 mA, see attached datasheet.
These have a dropout of 0,2 Volt. Measured output 3.28 and 3.31 Volts (HP3457)
The USB and the external power are connected by two diodes, so here its drops 0,5 Volt.
So one or the other or together can supply. The current i measured was 250 mA, with no load.
With to 50 ohm loads on 16 mA setting , the current was 280 mA.
The measured USB voltage was 4.7 Volts, depends on the quality of the USB cable.
With a regulated power supply, tested what the lowest Voltage possible was,
on the external power connector=> 4,5 Volt, then the outputs began to drop.
When i switched off the PC, the USB voltage was 0, and the external Voltage was still there and on the scoop
could see that the unit was stil working, so the problem of the lights going off,, is the USB data connection if the PC is off.
Took of the USB cable from the PC, and the lights were back on.
-
Hi Wim13,
Can you show a video of the problem or a photo of your setup.
Do you use hubs inbetween?
Does it work correctly if you unplug USB cable instead of shutting down PC?
Leo
-
It is not a big issue, because the unit works all the time. ( *** found the cause, see next post)
The PC is off, (= HP prodesk 400) see attachment. The USB cable is in the PC.
The external power is connected and turned on, there are no lights going on, and the GPSDO seems to do nothing.
But see picture, and the scoop image, no lights..... but the GPSDO is working oke, 10 Mhz E-10 on also on the HP5370 counter.
Then pull the USB cable out of the PC (still off) and the lights go on, still working as before.
If i then put the USB cable back in the PC (still off) the lights stay on.
So it seems only to happen on startup of the GPSDO when there is a USB cable in a PC that is off.
Now i know this, i take the USB cable out when not using is.
Will see if i can do some more test to trace the issue, maybe it has to do with the PC.
There are no hubs in between,
** an other remark, as you can see in earlier post,
in firmware 1.15 the serial number is gone to all zero's, and sometime in a week or so it gives the serial number.
-
Did some more testing.....and found the problem,
Put the scoop on the USB data lines to see what is happening.
See attached foto the red line to the plus data line is normal +3,3 Volts, with no connection, or a working PC.
When i connect the USB to the PC (still off), the data line is pulled to almost ground, the voltage drops to +0,3 Volts.
Then i took and connected a resistor of 1 k, and with disconnected USB cable...... ( the voltage drops to +1.3 V )
And then it happens......., the lights stay off when the data line is pulled down and the GPSDO switched on by an external power supply.
When i take the resistor away, the lights are back on, puting the resistor then again back, the lights stay on. Until switching the power.
So my PC shuts USB lines down when going off power.
And maybe the USB interface of the GPSDO signals an error on power up when the data lines are down ..???
Link to the datasheet of the PIC with USB interface see first post of this topic.
-
Hi Wim13,
Thanks for heads up, I will investigate.
USB standard requires that the data line pull ups (D+ for FS) are not providing any voltage when USB bus voltage is off.
I'll check how this is handled in the USB stack.
Apple had the same problem with this in 1990's.
Cheers
Leo
-
More info about: why are the leds off from my GPSDO, when USB connected and, my PC is off power.
Did read the datasheet of the PIC to see how the USB works, and analyse what can be wrong or not. :-+
for the other readers, see picture of the USB interface, and some explanation.
The D+ line (see picture) is set high to inform the host that it can work on high speed USB
When D- is set high, same that it can work on low speed USB
When both D+ and D- are low, that is called a: single ended zero situation, means connection ended or disabled, or reset.
When both D+ and D- are high, there is an error condition should not be possible.
What happens when the GPDSO is connected to my PC, when it is off with no power, the PC pulls down the D+ and D- lines of the USB.
That should be correct, it goes to the single ended zero state.
Checked some other devices, but not alle devices pull the lines down when off power.
So not all devices are implemented the same, on this point.
In the datasheet of the PIC of the GPSDO (see first post for the PDF), on power up it checks for single ended zero state. ( SE0 )
In my case, PC off power, the GPDSO detects the SE0 on startup and sets the register flag for that, what is correct. (chapter 24.2)
This situation is monitored until it come out of the SE0 state, during start up.
So what happens i think, as the two leds of the GPSDO are used to signal error conditions,
the off state of the leds, is also a condition to let it know there has been an USB error, because the unit stay working oke, without USB.
So to me it looks all correct, it has only to be applied to the manual of the GPSDO. 8)
But Leo is still looking into it.
-
Santa delivered my Mini so today I checked/adjusted my three elderly frequency counters. First up was the Optoelectronics UTC8030, good up to 2.4 GHz. I got this as a factory special deal at a Dayton Hamvention decades ago. It has the optional TCXO 10 MHz reference, but no oven. I had beat it against WWV some months ago, so it was still within a couple Hz. With the Mini it was trivial to tweak the oscillator to better than 0.01 Hz by watching the drift on a scope. Next was a Fluke 7260A. Again, with a 10 MHz reference. This thing lives in my garage and sees the heat of summer and cold of winter. I rarely use it but it still works fine. Amazingly, it was so close to perfect I didn't even open it up. Last was my HP 5334B. I installed an eBay ovenized oscillator in this and have never really tweaked it. With the Mini it was again trivial to monitor the drift. There's a trimmer on the oscillator itself and also a fine adjust pot on the counter board for electronic tuning over about 1 Hz. I set the control voltage to zero and used the trimmer cap to get it as close as possible, then the fine adjust pot to bring the drift to near zero. It didn't move for 5-10 minutes, which is certainly as good as it needs to be. Really pleased with the Mini as it makes this stuff so easy. I have some older DIY phase comparison stuff made for the 60 kHz WWVB signal, so it will be interesting to set 60 kHz and test that out. About the only thing I could ask for is to get it working with Lady Heather, though that's not high priority thing right now.
-
A question, if @Leo Bodnar is listening- There are no GPS signals in my basement because the rest of the house is in the way, but I do have a piece of RG-174 that goes from upstairs, to the basement lab. Is the Mini OK driving maybe 15 meters of RG-174, terminated with 50 ohms or unterminated? I'm assuming trying to run a GPS antenna signal down the same wire and putting the Mini in the basement isn't a good idea, but I don't have any experience with those tiny signals.
edit- Also, on the advanced page of the setup program, you get lat, long and elev. What datums are used for those?
-
I stand corrected and amazed. The Mini is running in the basement and picking up GPS signals with no problem. The sensitivity is remarkable. The setup program even runs on my ancient Vista machine down there.
-
@Leo Bodnar is it possible to add a period input to the program instead of frequency? For example 1mks, 2μs, 3ms (fractional frequency) .....
-
I have a Rev.F unit here broken with burnt input inductor from USB.
Is there a block diagram available ?
Regulator and MCU also dead. Probably lightning damage. Too bad I can´t get a PIC18LF25K50.
Edit:
LeoBodnar rocks !
I tried to replace the chip but then the firmware is not available.
But: they repaired it for me in short time. Very friendly folks !