You can mount an SMA socket on the GPS at the three connection points next to the small antenna plug

Hi, thanks for taking the time to make the video, you put the theory and practical aspects of the project across very well. I'm looking forward to the next thrilling installment!

@scullcomhobbyelectronics1702

8 жыл бұрын

+John Woods Thanks. Part 3 is uploading now.

Brings back memories training to be a ships radio officer many moons ago. Finished but stayed on dry land.Great series of videos and very enjoyable.

@scullcomhobbyelectronics1702

8 жыл бұрын

+Tom OConnor Thanks Tom.

Very informative videos thank you. The effect of the 'phase distortion' would be most likely cured if you put a BNC t-piece on the Rigol Oscilloscope, and add a 50 ohm terminator (ie resistor) on one of side of the Tee, and put your probe lead into the other side of the Tee. The ghosting is caused by an impedance mismatch in the cable, and so what you are seeing is the reflected wave bouncing back and forth along the test lead.

@RexxSchneider

2 жыл бұрын

That won't cure it. The "ghosting" is not an impedance mismatch, but the result of the synthesis of the 10MHz square wave from a 24MHz signal which is a non-integer division. The 10MHz is created by dividing the 24MHz by either 2 or by 3, so instead of 50ns high and 50ns low, you get a sequence of pulses that are comprise 3 pulses of 41.66ns interspersed with 2 pulses of 62.5ns in duration that average out at 50ns, making the frequency exactly 10MHz, but with an uneven duty cycle. That's what you're seeing on the scope. The solution is to set the GPS module to generate 1 MHz, which it can do precisely, then use a phase-locked loop running at 10MHz to drive the divider chain, diving by 10 at each stage. The 1MHz GPS signal is then compared with the output of the first divide by 10 stage to control the VCO in the PLL. That will give the full range of calibration frequencies as before, but with accurate 50% duty cycle square waves from each.

It's great to follow the design/build process of this project. Many thanks for your time.

@scullcomhobbyelectronics1702

8 жыл бұрын

+Eric Gee Thanks you are most welcome.

Again, a very clear video. I particularly appreciated the bit on filtering the square wave to get an approximate sine wave. Thank you for your very nice explanation of how to build up the filter and how it works.

@scullcomhobbyelectronics1702

8 жыл бұрын

+Paul Rautenbach Thanks for your comments. I think I will do a separate video on filters.

Superb. A pleasure to watch as always...and its educational too! Thanks for all your time and for sharing your knowledge with the rest of us.

@scullcomhobbyelectronics1702

8 жыл бұрын

+MrTom Thanks for your kind comments.

Thanks for everything that you have done in this project and for your time. I will get on and make the pcbs...........Berni

@scullcomhobbyelectronics1702

8 жыл бұрын

+berni ken Thanks. Remember when you print out the PCB layouts on to transparencies it is a reverse image and it is the printed side which then butts up against the PCB copper. This gives you a cleaner and sharper image to etch on the PCB. Good luck with your build.

MESSAGE FOR PIRATEMAN For some reason I am not able to reply to your comments. It could be due to the method you are using to post your comments as it seems to reject any reply. Below is the answer to your comment: Thanks for your comments. Building an LCR meter could be a future project, but as you say they are so cheap to buy these days. However, the process of building one would be educational. I will consider it for the future but at the moment I have a number of other projects planned.

Another great video, im training my ears in english and learn more about eletronics ... all the best from Brazil, NETO

@scullcomhobbyelectronics1702

8 жыл бұрын

+DJ Caverna Haroldo Bradaschia Neto Thank you.

Great vid!! Brings me back to school..and how i struggled with L-C-R filters..LOL!

@scullcomhobbyelectronics1702

8 жыл бұрын

+Robert V Thanks. School and college seems a long time ago to me now. I use to teach electronics in the 1970's - makes me feel old!! LOL

Excellent as always.

@scullcomhobbyelectronics1702

8 жыл бұрын

+Fracture Thanks.

Thanks for the explanations on the low pass filter arrangements.

@scullcomhobbyelectronics1702

7 жыл бұрын

Glad to hear you found it helpful.

Thanks for the hard work. I need to look for a good kit for building my own PCB's. This will be the first time I have made my own.

@scullcomhobbyelectronics1702

8 жыл бұрын

+Robert Calk Jr. Thanks Robert. You may like to checkout the two videos I did on making you own PCB cheaply. I used this method to produce the PCB's for this project. Links to the videos below: kzread.info/dash/bejne/rKOdq82wkaXcZtI.html kzread.info/dash/bejne/hHmiucSrgsrAf7Q.html

@robertcalkjr.8325

8 жыл бұрын

+Scullcom Hobby Electronics Thanks Louis. I've seen your videos and will watch them again no doubt. I meant that I need to find a good company here in the USA that has quality stuff for a decent price. I'll check and see what Jameco has.

Wow dude thanks! you have every thing laid out and vary understandable.. i am so making this!!!

@scullcomhobbyelectronics1702

7 жыл бұрын

Have fun.

A really great video.

Great work! I was looking for a Master Clock for my scopes and counters when servicing amateur radios. I found some cheap china modules working fine and even having a PPS output on the connectors. As the studio is below the rooftop, it works fine with the on-board patch antenna. If you use PPS output up to 1MHz signals are received perfectly but if you set PPS to 10MHz the receiver signal quality is dropping very low. I investigated that and found that the LEDs (my module has two of them) are driven by this signal and the signal is not a sqare but some sort of saw-tooth-ish. So driving two LEDs with 10MHz is putting some real ripple onto the complete boards power planes and loading the 10MHz to hard. I removed the LEDs and added some more decoupling capacitor to the Vcc and it looks far better now. even still without external antenna.

@scullcomhobbyelectronics1702

8 жыл бұрын

+Astralix2011 Thanks for your comments.

Yours is one of the better channels for the hobbyist electronics enthusiast. The detailed explanations you supply are often overlooked by others in this area of interest. I wonder if you considered using the original GPS and patch antenna remotely locatable with say an ethernet cable connection back to the divider and display electronics. At the frequencies concerned the losses in the coax are significant. By remote mounting you will achieve best signal to noise ratio. This is the way so called intelligent GPS antennas work. Thanks for the videos, looking forward to more.

@scullcomhobbyelectronics1702

8 жыл бұрын

+Rob B Thanks for your comments. I have found the use of an active antenna I illustrate works very well with good reception in all in door locations. It is also a very low cost solution. If you wanted to position the GPS module with its original passive patch antenna in a remote location and use an ethernet cable to connect it to the main frequency divider PCB you would need to use a GPS to ethernet hub remotely with the GPS module, which are costly and you also have the added cost of the ethernet cable. Plus you would need an ethernet adapter connected to the frequency divider PCB, again at extra cost. All of this would add a great deal of cost to the project which is not really necessary. The active antenna used in this project costs only about £4 (around US$6) and has a gain of 28 dB it uses a low noise LNA. As the Neo-7 GPS datasheet shows that the module operates very well in doors at an antenna gain of 15 dB then the 28 dB gain works very well with no drop out. The only benefit I can see from what you suggest is if you wanted to send the GPS data to many separate location on a network, then the extra cost could be spread.

@scullcomhobbyelectronics1702

8 жыл бұрын

+Rob B The other point I forgot to mention is the fact that this project has a wire soldered on the timing pulse output which gives us the 10 MHz signal used by the frequency divider network on the board. This 10 MHz signal could not be sent on the ethernet cable.

The firmware needs a fix ! At 19:33 the Sat display goes from 9 to 10 to 90 sats, so you should add an extra space after the number. Nice work overall.

Great video sir !

@scullcomhobbyelectronics1702

8 жыл бұрын

+Yash Kudale Thanks.

Great video as always! You could use one analog input on Arduino to measure the battery voltage.

@scullcomhobbyelectronics1702

8 жыл бұрын

+GADELHAS82 Thanks. Yes nice idea, I did consider having the LCD showing a battery check for a few seconds at switch on. I have a number of optional upgrades for this project which I will cover in later parts to this project. At the moment I have focused my attention on getting the project to give good accurate frequency outputs.

It is easier to get clean 10 MHz signal if you use LC circuit. 12 turns on 10 mm body (some 10 mm drill, remove after coil is made) with 1 mm wire. Parallel to the inductor is 250 pF capacitor. Whole LC 'tank' feed over 5 pF capacitor from 10 MHz GPS module output to this LC circuit, and at the same point you have clean 10 MHz signal. Since inductor is free standing in the air, it can be fine tuned by moving coil little bit. Take care about output impedance, since LC tank is generally high impedance circuit. You may use secondary coil with few turns, or solder tap on say 1/3 of the turns (4th turn from the ground end). Amplitude at this point will be lower, but LC tank circuit resonance will not be compromised.

Fantastic project, Just etched a couple of boards BUT still waiting for GPS to be delivered :-( Look forward to Part 3. MANY thanks Brian G0WZB

@scullcomhobbyelectronics1702

8 жыл бұрын

+Brian Burdis Thanks. Enjoy the build. I am working on Part 3 for this project.

Great video. Towards the end the display shows 80 and then 90 sats. That must be an error since there are not many sats overhead. My guess is it displayed 10, then 8 and the zero was not removed from the screen.

@scullcomhobbyelectronics1702

8 жыл бұрын

+Graham Wilson Thanks Graham. Yes you are correct at 33.51 the GPS module locked on to 10 satellites (just before that we had 11). I am finding that with the new active antenna it detects more satellites. When I wrote the software code for the Arduino I assumed that I would not receive satellites in double digits. I just need to make a minor change to the code to correct for this and blank out the digits first before displaying the new reading. I will provide updated code on the next part and may also include some additional features as well.

@andreasviehmann5805

4 жыл бұрын

@@scullcomhobbyelectronics1702 i am also serching for the new update Code for this Project to fix the uptate of more then 10 Sats with fix the 80 or 90 display , in your follow project with the keyboard i do not need this i only ned for this smaler project, can you update here the updated code for fix the 80 an 90 displaying ? Many thanks for to do this

@PowderMill

3 жыл бұрын

Super-Secret (normally HIDDEN from view) were detected by this amazing piece of kit. 😷👍🏻🤣 Seriously.. Great job!! Thank you .

Smashing design :-D I like the pluggable way you designed it, the modules can be unplugged for testing :-D I thought of the two basic circuits for filtering the hf content out too, the l.p.f Pi filter with all three component reactances being the same is something ive used on 27 mhz cb radio transmitters, a v.s.w.r meter can be used to get it roughly right for impedence. Also the tank resonant L&C parallel can become a filter. Im not sure how good the tank would be in comparison to the pi filter, ive never been into anything other than repairs. Your designing, thats a whole different ball game.

@scullcomhobbyelectronics1702

8 жыл бұрын

+zx8401ztv Thanks for your comments. For some reason Google put it initially in the possible spam list which I eventually noticed and had to tick to accept. Not sure on what criteria Google rejects comments! With regards filter circuit design, in the past I use to find that this was an area hobbyist new little about. It may be worth doing a video on filter networks and their design and uses if there is any interest in that. I plan to provide more details about selecting the correct value of components for the filters when I do the next part of this project in which I will focus on the output stages of this project.

@zx8401ztv

8 жыл бұрын

+Scullcom Hobby Electronics Im so used to parallel LC based vco's in synthersised radios that i forget about other uses, as i said the pi type i connect with filtering the output of an r.f pa or downmixer components, repair work doesnt need much, other than pattern/block understanding. Yes i suppose my brain needs a kick lol :-D

Thanks for sharing this. I was wondering if you would consider building a precision LCR meter project? I know one can buy one, but it would be very educational and useful. Just a suggestion of course. Cheerio

brilliant clear explanations of whats going on, thanks a million for the tutoral and instructions, when is part 3 coming online???

@scullcomhobbyelectronics1702

8 жыл бұрын

+Archie Versace Thanks. Hope to get part 3 done soon.

Using a make before break switch for the power section makes sense since losing sync from power down takes time to recover. But I wonder if make before break is a good idea for the frequency selection since that causes adjacent outputs to be momentarily shorted.

@scullcomhobbyelectronics1702

8 жыл бұрын

+Douglas Gilliland Thanks for your comments. In this project using a make before break rotary switch causes no issues so it is safe to use. Connecting adjacent outputs of the divider IC momentarily has no effect.

Hi first of all very cool Project :) but I wanted to comment on your measurements, I noticed you are using a switchable Probe on x1 so you can get only 6-8mhz bandwidth depending on your exact Type of probe, look it up in the data sheet, this might explain why the waveform looks worse with increasing frequency.

@scullcomhobbyelectronics1702

8 жыл бұрын

+jolo0011 k Thanks for your comments. You are correct about the bandwidth of the probe when used on x1 and this will effect the higher frequencies and shape of the waveform. I was using a slightly better probe but still the bandwidth would be better used on the x10 position. I have a number of different probes some better than others, but agree with you using the x10 option would be better. Will try and show next time. Thanks.

I am starting to believe that you have the best video channel on electronics projects on KZread ! I also have a question for you. I have been tweaking a programmable rubudium freq STD in order to get a highly stable 16.933 MHz for my custom made CD player based on a good old TDA1541 tweaked to the max. I was wondering what would be your recommendation on a circuit to upconvert a 10 MHz GPS reference to a higher freq ( ie 16.933 MHz). Is this even something you would consider doing and would it seriously increase the phase noise level of the resulting signal ? Thanks again for your excellent work !

@scullcomhobbyelectronics1702

8 жыл бұрын

+DJSolitone Thanks for your comments. Using a Rubidium Frequency Standard to derive 16.933 MHz for your CD Player seems a very expensive way of doing it. May first thoughts would be to use something like a Voltage to Frequency Converter IC such as the MAX038 or similar. The MAX038 is a high-frequency, precision function generator producing accurate high-frequency (triangle, sawtooth, sine, square, or pulse waveforms can be selected) with a minimum of external components. The output frequency can be controlled over a frequency range of 0.1Hz to 20MHz. Have a look at the datasheet.

@DJSolitone

8 жыл бұрын

Thanks for your reply. The idea I had behind the rubidium std was to have a programmable lab reference so that I could use on various applications in my lab (and also my sound studio where I manufacture custom sound appliances). I got the rubidium standard for cheap a few years ago, and managed to program it to the desired frequency, up to 20 Mhz. So I was wondering if such a think could be faisable with a GPS receiver like yours or something similar. I guess my overall question would be : How can you best upconvert a 10 Mhz signal to a higher frequency without injecting too much phase noise. I understand that it is easier to divide it and keep a good precision, but once you upconvert your base signal I guess that you get some loss in precision right ?

@scullcomhobbyelectronics1702

8 жыл бұрын

+DJSolitone I would imagine you would use a Block Up Converter which would simply use the GPS 10MHz reference in a Phase Lock Loop to lock a higher frequency Crystal or Voltage Control Oscillator to a precise frequency.

Can you tell me if the rising edge of the PPS has a fixed alignment with the rising edge of your 10MHz output?

Excellent project, I have started working on mine. I have been looking at Arduinos for quite some time and this project will make me get in it.I have a question on your diagram, why do you have jumpers drawn in the schematics, JP1-2-3-4-5-6-7-8I making a few changes on your schematics too, i'm planning to use a multiplexer ( I have a HCF4051 that I will try) controlled by the Arduino.

@scullcomhobbyelectronics1702

8 жыл бұрын

+dan3460 Thanks. Yes JP1-2-3-4-5-6-7-8 are simply wire links as the PCB is single side. With regards switching, I have been working on keypad entry of the frequency which will be covered in Part 3 of this project. A new single board PCB is also designed. Part 3 of the project is uploading as I type this so should be available soon.

A couple of comments/suggestions. Have you considered using a multiplexer to switch between the different frequency outputs? For example, you could use a 74HCT151 (or 74ACT151) to select from up to 8 frequencies for the one output, and then the mechanical switch would only have the control lines for the mux - no high frequencies running to the switch. Second, once source that I would recommend to you and your viewers that want to learn more about frequency standards and GPSDO (which is what you have assembled) is a mailing list called "time-nuts". You can find info on that and other sources of info here: www.leapsecond.com/time-nuts.htm . There are also some good threads on GPSDO over on EEVBlog. Also, there are quite a number of things that you need to watch out for in order to avoid introducing significant phase noise into the frequency standard. One of the things I've been told to avoid is high Q circuits, such as the circuit you were using to filter the square wave down to a sine. Finally, I have the same little antennas that you show in this video. There is also another one, its about 2-3x as large, which gets considerably better reception. The price is about the same. Anyway, thanks again for the videos.

@scullcomhobbyelectronics1702

8 жыл бұрын

+Matthew Humphrey Thanks for the info. There are a number of options I am looking at for frequency selection. I will cover more details in following parts of this project.

@mghumphrey

8 жыл бұрын

+Scullcom Hobby Electronics Excellent. I look forward to part 3.

Hi Mr Scullcom, Looking at you published schematic you appear to be running the Neo-7N from the 5V rail - am I reading this correctly. The data sheet for the Neo-7M states a maximum input of 3.5V? Great project, very much like the modular design. Barry

@scullcomhobbyelectronics1702

8 жыл бұрын

+g4rko Barry Thanks. I see you answered your own question in another posting.

How would I stop it defaulting back to 1MHz when losing lock? I'd require a constant 10Mhz, locks or not.

Every one I seen runs through a electronic frequency control input (EXTAL)pin on a ocxo 10MHz crystal then into a low pass 10MHz filter for a sinewave. Your probably picking up external noises from the wires around the gps module and the input voltages, lighting, ect. so shielding would probably help out. Might look at it with a spectrum analyzer to figure out all the issues. Look at with your FFT on your scope

@scullcomhobbyelectronics1702

8 жыл бұрын

+Martin P Thanks Martin for your comments. Yes you could also use this project to lock a 10 MHz OCXO and get a cleaner 10MHz sine wave out put. I have found that some of the noise is caused by the 48MHz fundamental frequency from the uBlox Neo-7. Adding some filtering on the 5 volt power line reduces this. Screening as you mentioned would also probably help.

@MartinPHE

8 жыл бұрын

+Scullcom Hobby Electronics, Be interesting to see what you find out. I'll order the chip sometime put it on my own pcb with my own components. I suspect some of the issues is with the pcb design and parts on the GY-GPSV2-NEO or V3. According to Ublox these things are really finicky with voltages though they don't go into detail about proper board design. Oh and you might try Power Save Mode since this might lower the noise and might be worth it to go through all the software and disable what you don't need.

@MartinPHE

8 жыл бұрын

+Martin P , found the design guidelines, thought I'd share. www.u-blox.com/sites/default/files/products/documents/MAX7-NEO7_HIM_%28UBX-13003704%29.pdf

@scullcomhobbyelectronics1702

8 жыл бұрын

+Martin P Thanks for the comments and info. Part 3 is uploading as I type.

Hey Louis. Quick question: How are you printing onto the whiteboard? :-) Your diagrams are so neat and tidy that they look printer or projected but your pointer doesn't leave a shadow.

@scullcomhobbyelectronics1702

6 жыл бұрын

gorillaau Thanks. They are all hand drawn, a skill I learnt from when I use to teach.

Thanks for sharing the PCB artwork. Would you be willing to share the actual KiCad files so that we can use them as a jumping-off point for customization and further revisions? I have KiCad and other software here in the lab, so it would be nice to have those files to work from. Thank you in advance!

@scullcomhobbyelectronics1702

6 жыл бұрын

Hi, Thanks for your comments. You can download the KiCad files from the link below. It contains files for both the main PCB and the Display PCB: www.scullcom.com/10MHz_GPS_Ref_and_Display_KiCad_Files.zip Regards, Louis

Hi Louis. The rotary switch is missing from the parts list! Discovered while ordering parts for this project :)

@scullcomhobbyelectronics1702

8 жыл бұрын

+Eric Gee Sorry about that Eric and thanks for bring that to my attention I will try and correct it. The rotary switch is a "make-before-break Rotary Switch with 2 poles and 6 ways" the one I used was manufactured by Lorlin Electronics Limited in their CK rotary switches range type CK1035. These are available from both Farnells (element 14) or RS Components Ltd. Cost is around £1.50. Farnells order code is 1123691 and RS stock number is 327-658.

@ericgee6585

8 жыл бұрын

+Scullcom Hobby Electronics Thanks

Very interesting video. Can you recommend seller for the neo 7. Most of the sellers i have found are in china

@scullcomhobbyelectronics1702

6 жыл бұрын

Hi Peter, If you use the following link you should find a few sellers. One is in the UK but also the most expensive. The ones fro China are the cheapest. www.ebay.co.uk/itm/1PCS-Ublox-NEO-7M-000-GPS-Module-MWC-APM2-6-Replace-NEO-6M-GYGPSV3-NEO7M-L30 Regards, Louis

@petergardiner2207

6 жыл бұрын

Thanks for your reply. I will take a look

Hi Folks: I am having trouble understanding how the signal from the GPS Satellites works to lock the frequency. I have done some Googling and reading and I am not getting what is the signal from the Sat is. Is it digital/analog? Also the Sats in their orbits are coming and going, so is red/blue shift an issue? Can someone provide a thumbnail primer on how this works. Cheers, Mark

@scullcomhobbyelectronics1702

8 жыл бұрын

+Mark Beeunas The GPS module (ublox Neo-7) has a built in oscillator (either Crystal of TCXO) which is used as the timing clock pulse for digital processing on the GPS module. This oscillator is also used to derive a timing pulse which is what we are using in this project. When the GPS module detects a satellite it receives an accurate timing signal which is then compared with the timing pulse and makes any corrections required to accurately set the output frequency of the timing pulse. As the timing pulse frequency can be set by us in the configuration settings, we set ours to 10MHz. It is this frequency which is then accurately tuned and locked and so we can use it as a frequency reference standard. Navigational satellites have a very accurate clock locked to both International Atomic Time and Universal Time (also known as astronomical time or solar time, refers to the Earth's rotation). Hope this helps.

To answer my own question! The Neo-7M is limited to 3.5V - the board on which it is mounted has a 5V regulator. Duh!!

@scullcomhobbyelectronics1702

8 жыл бұрын

+g4rko Barry Thanks Barry. It is easy to miss that the GPS module has a Low Drop 3.3 volt regulator on it, so we can feed the PCB from a 5 volt supply. The actual Neo-7M component in the centre of the PCB is what runs on 3.3v. The main PCB board is basically a prototype board.

For antenna connectors just be careful that rp-sma and sma have a reverse type of center pin (a Google image search can help you) img.hf-tech.org/items/fpv-ant-ra12s/02-l-Merkmale-SMA-RPSMA.jpg

@scullcomhobbyelectronics1702

8 жыл бұрын

+cyril muller Thanks for your comment. It was good to point out the differences between SMA and RP-SMA. The web page link you provided gives a good visual comparing both types.

21:34 90 sats? that's impressive ;)

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks.

@freelancer917

7 жыл бұрын

I'm sorry, that should have been a hint. There aren't even 90 GPS satellites in orbit, and you can receive maybe 20 at the same time. I think it's a bug in the LCD driver software, you don't delete the 0 in the second digit so if it displayed 10 and then goes down to 9 or 8, the 0 stays and it seems like 90 or 80. I like the project though.

@freelancer917

7 жыл бұрын

I just watched part 4 of the Series and the issue seems fixed there.

@scullcomhobbyelectronics1702

7 жыл бұрын

Thanks for letting me know. I smiled when I first read you initial comment of "90 sats" The LCD needed to delete the 0 in the second digit when it went below 10. As you point out this was corrected. Regards, Louis

Yuk! Stop biting your finger nails!