Sharing experience wit the time flies

[Skyraider]

Today is a great day because I completed my Time Flies kit !
This is by far the best experience in electronics since many years.
I would like to thank Tony for this great kit, for his support and his kindness...

It was not so easy to assemble and to fine tune it. I would like to share my experience and give some hints to others...

First thing to mention, I tried to assemble some SMD components with a hot air soldering system, but definitively, I feel more comfortable with my Weller soldering iron, and... a pair of 3,5x magifying glasses.

GPS Mouse:
- No difficulty with it. The only problem was the range of the 433Mhz emitter that was too limited (5m max)(I saw that during the tests with the Power controller unit).
- To improve the range, I removed the piece of wire that connect the 433 module to the antenna printed on the PCB (this was the wire described in the instructions: "Solder a piece of wire from the pad on the top right corner of the Tx module to the PCB - there's a hole almost directly below it:")
- And instead of that wire , I connected a 17,3cm insulated copper wire (I took a 0.5mm wire from an old transformer but a wire similar to the Power Controller antenna will be OK)
- I wrapped this wire around the GPS case.
- The range is now excellent (+20m)
I don't understand why the range was so poor with the printed PCB antenna...

Power Controller:
- No difficulty with it
- Just a warning: If you want to use the switch, be aware that it is connected to nothing. You will have to solder a small wire between the switch and the + connection toward the batteries. This is described in detail in a separate post in the Power Controller section.
- A great thanks to the Polyfuse that prevented a catastrophe when connecting the USB cable with the 15V and the GND inverted!!! (see below in the Levitation Base paragraph)

Levitation Base
- I had some trouble with the WS2812b leds in daisy chain: Vcc of Led3 (left) was not connected (bad solder). So the led received signal through the Din pin without Vcc connection and went out of work... I had to unsolder it. I ordered some leds through Aliexpress and i am waiting for them to be delivered to replace it. Of course, the unit can work without the Led...
- More serious trouble: My USB cable had a loose wire (data wire) in the micro-usb connector (intermittent connection depending of the position of the cable).After some investigations, it appeared theat the loose connection was in the cable and not in the micro-usb fixed on the board.
So I cut the micro-USB connector at the end of the cable and replaced it by the one provided in the kit.
I soldered:
- the data wire (blank wire)
- the GND wire (black wire)
-the Vcc-15V- wire (red wire)

When I tried this cable, I thought this was the end of my Time Flies kit :
- No power on the levitation base
- The Power Controller went on and off at high speed...

The reason is that the USB cable is VERY specific. It is mentionned in the instruction to be careful with that special USB cable but I thought that it was just because it provides 15V instead of 5V...
In fact, it is a "little" more specific:
- The 15V is coming from the Black wire
- the 0V is coming from the Red wire !!!
I soldered correctly the wires in the micro-USB and... to my surprise... everything was powered correctly. No components were damaged.
This is certainly due to the Polyfuse protection that saved the life of the components!

- As mentionned in other posts in the Levitation Base section, I also had some trouble with the capacitive sensor: When trying to fly the magnet, sometimes, the capacitive sensor displays a steady red light and the tone increases to the fifth step and never end: I suspect magnetic interference with the sensor. In that case, you have to cut the power and start again.
So I unsoldered the capacitive sensor from its location (above the PIC) and soldered it on the left side (as shown on a picture in the case assembly section). I described how to solder it in a post in the levitation base section. it is still not ideal (When struggling to put the magnet, it still provokes a steady red light on the capacitive sensor from time to time. But the situation is better, the sensor seems less sensitive to magnetic perturbation...). If anyone has suggestion to improve that, do not hesitate...

Clock Unit
- A little tricky to assemble due to the microscopic size of the SMD's
- During first tests:
--> One IN-14 refused to start: this was due to a MPSA92 collector not soldered: easy to fix
-->all tubes (IN-14, IN19) displayed several figures at the same time, as if several cathodes were activated simultenaously.
--> to fix that issue, I used some alcohol to clean the solder flux.
--> at that step, everything was OK but a remaining issue: for all tubes, when displaying a 5 (cathode pin 7) , the 6 (cathode pin 8)was also slighty lighted. And when displaying a 6 , the 5 was slightly lighted.
--> So I checked the resistance between pin7 and 8 of the tubes and I found... 10,2kOhm (instead of +30MOhms). After investigations, it appeared that it was tube #2 that had an internal 10,2kOhm resistance between pin7 and pin8. Fortuanetly, I had a spare IN-14, I replaced it and now everything is fine!

--> Last tricky part: Align the clock correctly above the Levitation base. I rotated the magnet as described in the instruction. It is not so easy to find the right position (I put a mark with a pen on the magnet to pinpoint its position and be more efficient in the successive tests)

Now everything is working perfectly, I am playing with the BlueTooth terminal from my Samsung to discover the different commands.
I am also reading the assembly code to learn how to use the "Blu Nose" commands: I will document them when I will be ready!

Thank you again, Tony, for this great kit !
I am looking forward to the next one

[leedawg]

SkyRaider that is so great that you finished your kit.

I am in the process of finishing mine things have been going pretty well, I ran into some problems with the base it just stopped working all of a sudden and turns out it was the crystal oscillator replaced it things are good there now.

You made mention one of your WS2812b LED's on the levitating clock did not work? All 5 on the top of mine were working but now the #6 LED is not working at all so am going to have to figure out why that is happening. Perhaps a similar problem you are experiencing.

I also noted you mentioned your touch sensor just stays on when you are putting the levitation unit over the base. I too am having this problem. It does not happen every time but does occur rather frequently. I have played around a lot with this to try and figure out why it is happening. I used to think it was magnetic interference to the switch as well. However, I am starting to think it is an amperage problem with the power supply. I notice the problem occurs when the levitation unit with the tubes pulls too much current it causes the base unit to drop voltage and then the touch sensor gets stuck on. I almost never have the problem if I place the levitation unit in place with one finger under it supporting it at about the right height while I use my other hand to center it then let it float. However, if i drop it into position with one hand the bounce up and down tend to make larger current draw and thus cause the problem I think. I wondered if adding more capacitors to the power coming in would help this or not.

Please post some pictures of your unit id love to see them. Mine is listing to the left heavily and I have tried adding weight to the hight side no go it just spins around so that the weight is on the low side. I tried the magnets on the base still no go does not help. Looking forward to getting mine all done hopefully in the not to distant future. Congrats again on finishing yours that is great.

Tony made a hell of a kit here I am loving it so far.

Oh perhaps im an idiot but I could not interpret the timezone command. I want to set it for pacific standard time which is -8 UTC. The directions mention 0x13,$PSU,6,4,17,6*** to set it for GMT -5, however, I do not see any correlation to -5 GMT here my next step was going to be going through the assembly code to look into it. I also am not sure what is meant by the RGB value in the led command. It does not appear to be a hex value or decimal code so not sure what the last two digis represent the example is $LED9,B,5*** so this changes led 9 to blue I presume however not to sure I have modified the B and 5 and get different results. Most of the time just changing the brightness of the LED.

I also did not understand what you were talkin about with the USB cable. I, in fact, destroyed my original USB cable by trying to clear some solder out of the socket. managed to get the solder out but the cable was the sacrifice. Anyway I just purchased a new micro usb cable and plugged it in and everything is working, well minus the issues I mentioned above but I figured that is not due to the cable or I would not have any of it working. Would like some more elaboration on what you were talking about.

Any way look forward to hearing your take on some of the issues im facing. Tony has been super helpful, guiding me in the right direction.

Thanks again.

lee

[Skyraider]

Hi Lee,
thank you so much for your comments !

When you write "...but now the #6 LED is not working at all..." you mean the Led under the right tube or the last Led (the one under the clock on the right side)?
If this is the led under the right tube, it is strange because if this Led is broken, it would not be able to reshape the signal to transmit it to the last LED (the one under the clock on the right side)... Did you try to send a LED command to it? (e.g. 0x13,$LED6,G,5***). If this is the last LED, you should check the different pins. On my case, I had to connect a oscilloscope (picoscope) to ensure that the LED was really broken...

To answer to your GMT question:
- On my case (i live in France) I am at GMT+1. So the command is: 0x13,$PSU,6,4,1,6*** the 1 means GMT+1
- On your case (if i understand well Tony's instructions, as you are GMT-8 you have to put N=12+8 = 20 (because it will compute 12-N to find -8)
- So the command is 0x13,$PSU,6,4,20,6***


Coming back to the LED command : You have to know that in the asm source code, each LED has 3 registers: one is for the Red Brightness, the other for the Green Brigthness and the last for the Blue brightness.
So with the command, you will design the Led number (e.g. 0x13,$LED6,G,5*** : the 6 means led#6), the register for that LED (e.g. 0x13,$LED6,G,5***: the G means that you will adjust the Green brightness for this led) and the brightness for the choosen color: (e.g. 0x13,$LED6,G,5*** : the 5 means that you want Green color for led6 with a brightness = 5)

With that , you can adjust the R, the G and the B brightness for each LED and then generate many different colors. (e.g if you send 0x13,$LED6,R,5*** and then 0x13,$LED6,G,2***, The LED#6 will take an Orange color.)
Note that in the WS2812b datasheet it is said that you can adjust the brightness from 0 to 255 but here, the maximum possible value is 7 (I checked that in the code): this is to limit the current of the clock. (if you put more than 7, it will be accepted but the code will do a logical AND between your value and seven: that will keep only the 3 first bits of your initial value).
I suspect that there is a command to send all the LED values at the same time (LED packet). But It doesn't work for the moment...I have to dig a little more

Related to the USB cable: I should have done like you: replace the cable. But i chose (bad choice!) to cut the micro USB and solder another one: the wires in the USB cable are: red, black, blank and blue. Blank and blue are for data and Red and black are for power. normally red is for positive (+5V) and black for 0V. So i thought that the +15V was on the red wire and this is not the case here: The +15V is provided by the black cable and the 0V by the red cable...

I posted some pictures of the kit (you will see some other Tony's kits in the backgorund!)

Best regards

[leedawg]

SkyRaider, Thanks so much for your reply been super busy this past week and managed to get in a car accident Friday (lady crashed into the back of my car going 70km/h) any way I finally got back to the clock project.

Your explanation was immensely helpful, I do not know assembly at all. I have only programmed in C++ and my only microcontroller experience is with AVR's and thus using AVR C to program them. After you explained the time zone thing as well as the LED's it makes sense. Interestingly I had to change the offset to 19 not 20 to get the correct time due to the fact I forgot we are on daylight savings time here in California right now. But what you explained worked very well.

As for my broken LED yes I was referring the LED on the far right top side if you are looking at the display and it is LED6. So I unsoldered all the pins for the socket then removed the LED and replaced it with another LED. Now it works fine with the rest of them. Not sure what was wrong with the original one but got the problem solved by replacing it.

What I am fighting now is getting the thing to float level over the base seem to endlessly be adjusting the rotation of the magnet and trying weights on and off. So what I have done now is place one of the small magnets I pulled off the coil on the magnet of the floating part and that seems to have a pretty good effect on the floating angle. So experimenting with that trying to get the orientation to stay facing forward.

Does yours tend to rotate back and forth as it is hovering or stay very stable in one position after it settles down?

I have not finished the case yet for the base but have noticed while the relay portion is floating if it turns 90 degrees the IR receiver stops picking up the IR signal being sent from the IR leds, so If I put my hand near one of them thats enough light bouncing off my hand for it to pick it up again. How does the IR signal make it to the clock with the solid top cover covering the entire base unit?

It appears as though yours is working are there spaces in the slats and enough of the light emitted gets bounced off?

Anyway your clock looks great hoping to get that point pretty soon with mine. Thanks again for all the updates and clarification on your build I found it super helpful!!!

And I still am dealing with the touch sensor getting stuck on when either the relay crashes or I am trying to put it in position to float over the base. I may end up moving it as you suggested to the left side.

Lee

[Tony]

You'll need to rotate the magnet in the floating section slightly to change the direction it faces, once set up it should be static. The small magnets can help with altering the direction and any tilt once it's close.

The top cover is transparent to IR.

[Skyraider]

Hello Lee,
sorry for my late answer (i was out with no access to the forum).
I hope everything is OK for you and that only the car suffered for the crash.
For me too, it was difficult to get the clock aligned with the base: I had rotate the magnet fraction per fraction until finding the right position. it was not so easy, because if you go beyond the right position, the magnet can suddenly rotate 90°... By doing all these adjustmenst, be careful not to break one of the coil wire!
In my case, the clock floats horizontally so I had not to worry to compensate with additional small magnets. I had only to concentrate on aligning with the base.

You mentionned that you loose contact with the iR leds when the clock is rotating: this is normal, you will catch the signal only when the clock is aligned +/-40°with the base.

I made some finition on my clock: trim by 2mm the tubes (the wires were too long) and polished the paxolin with olive oil... Look great now !
Best regards
Emmanuel