Yannick99

Hi UR8IP, good work :)

Yes in corner, S= number of sat. You have never seen more satellites than 5 ? Pretty sure it's your problem. I dont know in u.k. how many sat you can receive. Maybe you can try to find a better antenna ?I know that pulse appears after 3 satellites. But i do not have any scientific explanation but it's like the pulse isn't accurate when you have few satellites.Last update is 1.54. Refresh the page i uploaded this morning.

Hi Gavros, it's weird. Did you saw my last instructable here:https://www.instructables.com/id/3D-Printed-Box-Gp...I'm using a 131-100 like you. And it's working. The frequency is moving a bit more than my double ocxo but it's very ok. Take a look.1: pwm directly in ocxo vfc is ok.2: Version 1.4 is old but it was the classic one with 6 phases. It's not the problem.3: By experience those antenna are trash, but if you are directly outside or near a window and you have 9 to 10 satellites it's normally fine. 4: How many satellite are you seeing ?5: Be sure than your power supply is large enough. See the last messages below. It was the problem of Simon. 131-100 take 600 ma alone.So my advice: be sure to have more than 9 satellites, a good power supply and if you can, install version 1.54. 1.4 i…

Hi Gavros, it's weird. Did you saw my last instructable here:https://www.instructables.com/id/3D-Printed-Box-Gp...I'm using a 131-100 like you. And it's working. The frequency is moving a bit more than my double ocxo but it's very ok. Take a look.1: pwm directly in ocxo vfc is ok.2: Version 1.4 is old but it was the classic one with 6 phases. It's not the problem.3: By experience those antenna are trash, but if you are directly outside or near a window and you have 9 to 10 satellites it's normally fine. 4: How many satellite are you seeing ?5: Be sure than your power supply is large enough. See the last messages below. It was the problem of Simon. 131-100 take 600 ma alone.So my advice: be sure to have more than 9 satellites, a good power supply and if you can, install version 1.54. 1.4 isn't the problem but with 1.54 you have only 5 phases instead on 6 and different bugs are fixed. You will also have classic mode (with jumper) or quick mode. My 131-100 is working with quick mode and classic. Try both.Also 131-100 is a simple ocxo. The ambiant temperature need to be a bit more stableLet me know

Hi, sorry for the delay. For t3 you can use any 230v/12v. Just maybe check for the pad layout if hole are in same place.T4 is different. It's a current sense. Probably some other will do the job. But i can't recommend any. The spike on uC side must be high enough to fire the start.Check on ebay i saw some shipped worldwide.

Glad to read that. Like i said, dont think you need an op amp. Remove it. Will be a bit more accurate. Enjoy :)

Your are correct. pwm max is 5v. You need 6v. Amp x1.2 or so is needed. But like i said, your current pwm value is 6a98. Double check the gain of your amp.But it should be works, i dont know why your frequency is moving around.

Just did a test code. It send pwm 0 and display the frequency on 10 seconds gate. It send pwn ffff and display the frequency on 10 seconds gate and restart.https://1drv.ms/u/s!AnKLPDy3pII_v2E4Gr37X3LmmYp9?e...

Ok. As i can see you didn't need an op amp. And the vfc voltage from ebay seller is maybe good. But you have a stability problem for sure. Is your transformer is large enough ? Voltage is good ? Doesn't move ?

Have you tried 1.53c (classic mode). See step 11 you will a picture where to install the jumper.

Hi Simon, It's not the problem. I have built on friday another gpsdo with a 5v ocxo. I put the gps module just above the ocxo and it wasn't working well. A bit like your results. I had 7 sattelites. So i tough that maybe the 10mhz square wave was too close of gps module. I just forget it and go to a weekend and I came back from camping tonight and guess what. It works fine now. Difference is i have 11-12 satellites.So all is goodI searched your ocxo and yes the vfc is around 6v. If i look your eeprom picture. The pwm config is 6A98. With 6A98 the pwm output voltage is around 2.08vSo your amp isn't 1.2x but 3x right ?Maybe you could try to lower the gain of your op-amp. The pwm will be higher but more accurate. Exemple, if you catch noise it will be amplifier x3 on the other side.

Op amp of 1.2x looks not too bad. It should be working. Have you put the gps module near from OCXO ?

Well done. very good work :)

Hahaha it happen to me yesterday. All night with door wide open :) It's a good idea. I'll check what i can do for that.

Done... check GarageDoor2.rar You will see closingTime = String("02:00") Just put your time to close in UTC format. It's very basic and you must reprogram the device. But each day at same time it will check the door and closi it if it is open.

It's strange, as i can see, you is in run mode but frequency have drift of 50 hz on 1000 sec. It's huge. In classic mode, pwm is always plus or minus 1. It's 76uV of change in 1000 seconds. Now it's hard to explain why you have a very large step. Maybe the vfc cable is capturing noise ?

Cart is now fixed to 4x 4.7uf thanks for the feedback. Some parts was cheaper on buy by 10x so yes some part are too many. It's up to you to adjust the cart before. Bye

Hard to tell just like that.

Le led gps pulse clignote, c'est normal oui. Pour le fréquencemètre, je n'ai pas l'intention et le temps pour le moment... Peut-être un jour :) Bye CharlesYannick

Pourquoi passes-tu par un ampli ? Pas besoin.https://gerrysweeney.com/tag/ocxo131-191/Regarde ici. le vfc du 131-191 est 0-5 volts. Tu dois faire exectement comme dans mon schematique.

Donc tu dois lire fréquence inatteignable ? Sinon ça marque quoi ?Au démarrage quand tu lis sans config 50% utilisé. Tu devrait être au alentour de 2.5VSinon tu as un problème avec soit le uC lui même ou la charge de ton ocxo est trop grande ou defect qui abaisse le pwm a 0v.

Que veux-tu dire par plus de commande pwmw? Tu as 0 volts? Ça bouge pas ?Ça affiche quoi ?

If leds 3 and 4 are off. You haven't reach run mode and you dont have any pulse because counter led should be always on. Or the leds are defectives.Also a drift oh 4 hz is almost impossible. Something is wrong.

At 3.6V it should be working.

It depend of the range of the VFC (EFC). I dont know for the ENE3311E. I tried to find the information but found nothing about this. What is the vcfc voltage at step 3 ?Try version 1.52 classic mode (with jumper)

Your Welcome. Glad you enjoy! Yannick

Welcome, glad to see everything is good!

Oui c'est une bonne idée une version française. Et je dois dire que ce n'est pas très long a faire, il n'y pas grand chose a traduire :) Le problème est que les afficheurs hd44780 ils ont 2 versions. Européen et Asiatique. Donc si je met les accents (éà), ils appariaient en chinois sur tous mes afficheur vu que je les ai acheter sur ebay.Je pourrais faire des caractère spécial en utilisant la ram de l'afficheur, mais pour cela je devrais changer plusieurs chose dans le code.J'ai donc fait une version française et je n'ai pas tenu compte des accents.Aurais tu une idée de quoi mettre à la place de RUN ? J'ai pensé a CRS pour course. Ca doit prendre 3 caractères et remplacer "RUN" pour mode run. Pas évident.

Hi, look step 14. Red is 12v Black is GND, white is the 10mhz out and yellow is Vfc.I checked your link it's a nice gps module. You can use it. But a cheaper one can do.

Hi, you click on schematic, and you click on the arrow at bottom left (download original).

Wow well done, that's i like to see. Now it's normal. Good to know for the dc/dc converter. I think that the most important thing is the ambient temperature and not moving the device (in height). Best Regards, Yannick

Great glad to hear that. I added anyway a code to reset all atmel memory at startup. You had a hardware bug but maybe if the software was erasing memory it was still working fine. Thank to you to have point me this problem out. With you video, and coordinate, I know now you are leaving in Slovakia 😊. Bye from Canada.

Hello, me too i do not wait 15 min when i test. I double checked on my side. i was in run mode for days. Dont remember if i had press button or not. But i disconnected the antenna, i saw self running. Reconnect the antenna. I turn off on, press button to pass. After i saw ph4, 200s (because i had a config found). I removed the antenna and saw no pulse self running.So on my side looks like watchdog is working fine if i bypass the waiting time. My fuse are extend:FF high D9 Low E0.Double check my code. The waiting time is a sub-routine when this one is done or bypassed it came back and do all next instruction including watchdog. very strange problem.

Yes you understand well. Now listen, i think i know your problem. It happen to me too sometime. When you turn off and on your device. You must wait to capacitor discharge. Also turn off your stk500. If not, the are some voltage keeping memory active and false, put bugs when back on. In software, i'm clearing all register to 0 but not not memory.So be sure you start the uC "clean". Try thatAlso my email is agaguk_99@hotmail.com it will more quick and better.That's said, i never think what flag exactly is in trouble here. I tough i was clearing all flag but maybe i miss one.

In fact, with avr i had 2 choices. 1 or 2 second watchdog. I can't choose 1 second it was too quick and happen when uC was counting. All watchdog counter is different. uC use another oscillator (128khz). and this one can change a bit. So i'm using 2 second i had no other choice between. So with 2 second it may be take 1 pulses to discard or sometime more. It's problematic when the satellite number is low. I'll tried to find a way to do it different but so far it is the best i could do. Try to jump more than 1 pulse you will see a self running message appears.

Yes i used little step for phase 1 to not overflow the pwm on ocxo where good adjustment is near 0v or 5v. When pwm overflow you have unreachable message. And on 1 second step is no big deal. Glad to see it's working for you. Bye

The watchdog overflow discard the count when a pluse is missing. If you read 1023 it's because you are running at 10,000,001,023 hz. I think you are very offset. 1023 = 3ff (ff) and 1024 = 400 (00) uC is looking for the end of this number so it's stick on 1023 1024 1025. Also sat=5 is a bit low. It's means than sometime it can be lower than 3 and now you will have intermittent problem and miscount

Thank you Igor. Keep in mind. Classic or quick version, at the end. It's the same. +- 1 on the 16 bit pwm when frequency is FF or 01. But for quick version if difference is >1 the correction step will be a bit more. Normally you do not have step >1. So with +-1 (15,3uv) it's the tiny step i can go to stick on and it did. I can't do an average. I tried. My result was worse.that's said, when you is in run mode, your good.If you tried a voltage divider like you want to do. It will work. You may have good result but if temperature is moving faster you may have a lack of adjustment. Also the gate is always 1000 seconds. To work with your idea, gate sampling should be higher than 1000 seconds. It's not an easy task to shoot and look more than 1000 seconds if you have hit or miss the targe…

Thank you Igor. Keep in mind. Classic or quick version, at the end. It's the same. +- 1 on the 16 bit pwm when frequency is FF or 01. But for quick version if difference is >1 the correction step will be a bit more. Normally you do not have step >1. So with +-1 (15,3uv) it's the tiny step i can go to stick on and it did. I can't do an average. I tried. My result was worse.that's said, when you is in run mode, your good.If you tried a voltage divider like you want to do. It will work. You may have good result but if temperature is moving faster you may have a lack of adjustment. Also the gate is always 1000 seconds. To work with your idea, gate sampling should be higher than 1000 seconds. It's not an easy task to shoot and look more than 1000 seconds if you have hit or miss the target.

Thanks LukGasp

Salut Charles, comme tu as pu le voir tu ne peux pas faire ce projet avec un quartz ordinaire. Celui ci va trop varié et jamais tu réussiras à rester a 10mhz +-.001C'est pour cela que ca prend un OCXO. Cela dit je n'ai pas de liste officiel a te donner. Par contre sache que le pwm fait varie de 0 a 5v le vfr de l'ocxo. Si celui ci supporte 0-8v tu devras probablement le matché avec un op-amp pour compenser. Si c'est le contraire, exemple de 0-1v, tu devras faire un diviseur de tension. Regarde les fiches technique. Le code est fait pour varié de 0-5v avec un pas de 2 hz par volt.

But it's almost the same of now. When run mode is found and value is saved in eeprom. GPSDO always starting at this speed and go directly in phase 4. But i understand what you mean.

I like to see other people project. Good work :)

It's between 2-3 secondes of stabilization. Do you think is enough ?

1.51 run my new algorithm and doing larger step to go in run mode. But not compatible with all ocxo.1.5 go slower but work with more. dont think that your ocxo is in problem. Just compatibility matter.Working on 1.52. it will be 2 version. The classic and the quick.. Bye!

Great! :)

Hi John, i had some comments from others with same problem of you. 1.51 is doing large step to reach 10mhz quickly. But if your ocxo isn't like mine (cmac 2187c) the vfc step adjustment is maybe different.So i'm currently working on version 1.52. 2 versions will be available. The quick and the classic. In your case the classic will work as before.Now you can install 1.5 https://1drv.ms/u/s!AnKLPDy3pII_vWPth5qYJp0cd0mZ?e=3ZT1Op

Maybe you could install a tension divider ?http://www.ohmslawcalculator.com/voltage-divider-calculator

You need to this with safari itself: I'm working at 300% on iphone 8plus. And your phone will remember the size for this shortcut.Launch the app and navigate to a website, then tap the "aA" icon in the upper left corner of the screen. The text zoom options appear at the top of the Web View dropdown menu – tap the smaller A to reduce the zoom percentage and the bigger A to increase it. It's that simple.

Router port is open you mean you did a port forwarding to the garage door ip adresse ?

I cant find the schematic of your ocxo but this is what i think. Your ocxo have a different vfr than mine. So with the new finding algorithm, the uC always miss the target. For you, you need to use the original algorithm ie: v1.5. https://1drv.ms/u/s!AnKLPDy3pII_vWPth5qYJp0cd0mZ?e=Q0Q97j

Hi, what ocxo you are using ? Never seen your error before as i program in different way. 1.51 use same fuse bit as before.

Ok good to know. Thank you

So if output is better. You have 10mhz to 74f04 pin 6 and 12. To pin 6, it is a nice 0-5v square wave ? Maybe the output to pin 12 is nicer but at pin 6 it's maybe Crushed. The uC load is maybe too much for the 74f04 or something like that. I dont have any 74F04 here so i cant test. If it's really not working i'll add a notice to avoid the 74F04. Keep me in touch

Probably the uC doesn't oscillate with the 74F04. Do you try to adjust the potentiometer? This pot is for this purpose.

Your welcome, thank you for your comment.

Very nice work, i like the idea of dividing frequency to have a lower one. I already use a 74AC390 to do this but externally. I'd like to do a similar device to add externally to my current gpsdo. Or i could use a Si5351C. This little chip can do any frequency with a good know frequency. But need to be programmed. I could use another atmega with keypad to do that.... Maybe a future project with a 4 line display :)Again good work.

Wow, very nice design. Your welcome. And again good work.

Wow, beautiful. Congratulation it is not an easy project to do. I like very much your design. Good work.

Pdf are working. I'm tring to upload a .zip file. Yes tried to logoff. Tried another browser. I'll try .rar instead

Hi, you had 1.4 and it was working before 1.5 or you building a new device ?

Good work... you are in run mode. It means that you have met at least one time 10,000,000.000 count. But now (on picture) you have .011 It's maybe the simple ocxo instead double, temperature changing too fast or simply the gps pulse itself. When you think about that. It's a huge challenge to have the last 1000 second pulse directly on the last count of 1 billion. But as you can see on picture of eeprom file. It's suppose to be very stable.

Hi, should be working. Vfc is in range . I have the same ocxo on my 5v version i tried and it was working fine. Wait more longer. Is the moving frequency make sence ? I mean now i see 78 but it should be going down over time...

... you can restart. You are not lucky. Normally it's faster than that. Probably the frequency has changed after phase 4. Did you bypass the 15 min waiting time ?

Your welcome om4ik. Happy holidays to you too :)

Version 1.5 is ready. It correct a counting bug. Sometime the frequency wasn't count properly. was a bit too high exemple 10,000,000.006

Version 1.5 is ready. It correct a counting bug. Sometime the frequency wasn't count properly. was a bit too high exemple 10,000,000.006

Version 1.5 is ready. It correct a counting bug. Sometime the frequency wasn't count properly. was a bit too high exemple 10,000,000.006

Version 1.5 is ready. It correct a counting bug. Sometime the frequency wasn't count properly. was a bit too high exemple 10,000,000.006

Nice looking and great wook. I know what i'm talking about :) https://www.instructables.com/id/Nixie-Clock-YT/

hard to find just like that. I can give you the code: https://1drv.ms/u/s!AnKLPDy3pII_ul12b4-Mcu-K8IIG?e...The device trig on int1. You could try to rise the starting voltage. Now it's 900 (384). more the value is high less the voltage is. So try 850. On my side with 120V the current is higer than you so i dont need a huge number. You can open the file with atmel studio 7

Very good design, good work. Last week i fixed an amplifier rotel with this probe. Was probing at output and i found directly where the problem was. Very handy to have. Your welcome.

Hi Gerry. The licence is at the beginning. CC BY-NC-SA Published Nov 2nd, 2017Stats

Ah, just seing you have upload a third photo. It's better :) Good work

Your welcome... But how can you be in run mode and have an error of 2.55 hz. It's huge. And the satellite led is on but no sat is displayed! Maybe the led board is upside down ?

https://1drv.ms/u/s!AnKLPDy3pII_u1KGRpvJapUmQti1?e=58aFzS

Hi, did you receive my email ? returning me back 2 times ?

Hi, i just checked and yes it is available. have you click on nixie tube.zip ? this is the link. https://cdn.instructables.com/ORIG/FF0/9SIQ/JW6NZSZ6/FF09SIQJW6NZSZ6.zip

Hi, if you have 12v at tp1. it's because there aren't oscillation at irf644 gate or the mosfet itself it burn. I dont know if you have an oscilloscope but what do you see at pin 4 of 7404 ? if not what voltage are you reading ?

At 0,8 volt you should be have around 30v at tp1. For me i have 2.1v for 180v. When you do the hvps calibration, when you hit + the voltage. Mosfet gate voltage rise (7404 pin 4). The coil L1 is ok ? You can sens me email instead. Instructable have a huge delay. agaguk_99@hotmail.com

Enjoy! Yes double check is a good idea, sometime i can do mistake ;) but should be ok. I choose not expensive part. Some was more cheaper to buy 10x so...

I did a digikey downloadable cart. check in parts section.

Hi :) Thank you for the gentle comment. I know exactly what you are talking about. Those zener diode are for protection and can burn anyway if bad thing happen. It's why Paul said to install those one in a easily accessible area if you must replace it. It's regular zener diode but a few watts will help. This should be fine 1N5352BG 5w zener diode 15v or 1N5929BG 3 watts. You can even install 500mw but, you will have more chance to break it if a high voltage happen at the probes. So why not install 5w :) For downloadable cart i'll see what i can do.

Good work with your counter. I didn't know the reciprocal way. I like that. I'm currently building one for frequency of 100k to 500k on lcd display. Thanks for sharing.

Hi, the source code is an assembler (.asm) all comments are in french

I sent you .sch If you do this, you will love your clock. Let me know!

Oh Nice. well done. Thank you for the gentle comment.If you find what was the problem tell me. Maybe helpful for other one.

Ok you means PAD sorry. Because PD6 PD7 are uC pin. Sorry my error. Now for your problem. If both LED are responding it's good. The program is ok, the current sensing is ok. By default, LED 2 should be turning on delayed after LED 1. No need to connect vacuum. it should be working with no load.For connections, to be sure, Pad 6 3 1 to main, it's ok. Pad 2 and 4 to the outlet. Dont forget to cut the connection between both outlet on the hot side. And the 2 others side of the outlet go to neutral. I saw 2 white wire on your outlet, those cable need to be together. Dont need to cut the connection between outlet here. Only hot side need to be cut. Probably your purple cable are hot. So on this side cut the connection and not on the other side. If you cut the connection on both side just jum…

Ok you means PAD sorry. Because PD6 PD7 are uC pin. Sorry my error. Now for your problem. If both LED are responding it's good. The program is ok, the current sensing is ok. By default, LED 2 should be turning on delayed after LED 1. No need to connect vacuum. it should be working with no load.For connections, to be sure, Pad 6 3 1 to main, it's ok. Pad 2 and 4 to the outlet. Dont forget to cut the connection between both outlet on the hot side. And the 2 others side of the outlet go to neutral. I saw 2 white wire on your outlet, those cable need to be together. Dont need to cut the connection between outlet here. Only hot side need to be cut. Probably your purple cable are hot. So on this side cut the connection and not on the other side. If you cut the connection on both side just jump a cable on the neutral side (2 white cable together). For your fan. probably ok, try it. But you realy dont need a fan. You will add noise and dust for nothing. The thing doesn't get hot. And i saw your heat sink :) You will ok.

OK2 and OK3 are optocoupler. So the load and the LEDs are completely isolated. Trust the led. At start, led1 is dimmed. led2 is off, this is normal. Note that LED are reflecting the output. Led1 dim = output dimed. Led1 full on = output 120vac. Same thing for led2. If led1 is dimmed at start it's because we need a bit of voltage on the saw but not 120vac. Just a bit for starting.If you are able to see led2 turning on, (slow or not) and you dont have any current on output 2 you have a problem. Maybe triac ?Is led2 turn on sometime ? When ?