HDDClock - clock from hard drive

Thanks for the link
The question came from the fact that I am still a layman in this field.

Hello mate
can this be done with a pic 18f series or 30f series microchip microcontroller , if so could some one post the code please

Hey.
I am looking to this project and I have a question about diode control.
The plate is spinning, the slot changes position and now ... at the same time you control the color of the diodes, i.e. the slit moves a more and more new color,
whether
you register a sensor that you are at point A, turn on the diodes red and immediately turn off. A red strip was created. The platter makes a full turn, now it is at point A + 1 (you calculate the moment you find it there knowing the speed of rotation of the plate) and turn the diodes purple - the color got through the aperture at point A + 1, you turn off the diodes, the plate must do again full turnover, etc.?

I hope that I have clearly described the two ideas that came to my mind.

Second doubt: why did not you use the electronics from the original disk to spin the plate? The revolutions would be very well stabilized and you would save on additional circuits, connections, etc.

The individual colors of the diodes are connected in series and all such series to one uC port with the transistor as a key, right?

Greetings,
sniff.

Diode control is carried out as follows:
- the rotation of the plate is divided into 240 steps (time slots)
- in each time slot, fast, 5-bit, three-channel PWM is fired
- the fill factor is loaded into the channels for red, green and blue respectively
- in this way, any 'time' color is generated in each time slot.

The original electronics could not be used for two main reasons. First of all, the engine simply did not start. This was probably caused by the fact that the electronics first checked whether the disc contained everything that was needed, and then it started with the revolutions. It seems logical, especially if the head of the disk were damaged. Then the rotation of the engine could cause serious damage on the plate's surface. But even if the engine was to start, the original electronics would have accelerated it to such a speed that the vibrations would be unacceptable. My plate rotates at a frequency of about 50Hz and this is a sufficient value (almost imperceptible vibrations and the image quality will be better and it will not be any better). I omit the fact that with original electronics it would be harder for me to design outputs for power and communication sockets.

When it comes to connecting diodes in a ribbon, all diodes (or rather colors) are controlled in parallel by n-mosical keys.

Thanks for the answers.

Do you implement PWM by software or hardware?
In the schematic I can not see any diode control system only from the B port through the resistor switch the key that turns on the right color, right? So programmatic?

I have not checked yet whether my disk without heads will start. Although I was hoping that spinning a plate would arrange the disk electronics for me. I have Segete, so he will probably behave like yours.

PWM is hardware, i.e. I use one of the microcontroller peripheral modules.

I'm sorry to bombard you with a hail of questions, but I would not want to commit a stupid mistake on the project.
I hope I will not unduly strain your patience if I take a second?

I connected my molex plug-in plug from my PC's power supply with the slot already cut out ... the conclusions brought me into the abyss of despair.
Although the drive takes off, it vibrates like a very strong vibrator in the phone. The watch would wander around the table.
After about a minute, the disk electronics stop the plate.

So I will not run away from the disk engine driver.
This dedicated driver you used was TDA5140A (please, pay attention to the order of numbers: 5 1 4 0)? You write about such an assembler on the first page of the thread. The schema.pdf file is TDA5410A, only google spams some datasheet, I suspect that there is a typo on the schematic?

Your engine had 3 wires on the disk, right?

Scalaczek is terribly difficult to access, but after checking the electrodes, diodes and others for a day, I come to the conclusion that the self-built, simple BLDC motor controllers are the first to take up too much space. Second: they give unstable turns. They will stop every now and then, start to spin again, etc.
I understand that thanks to the control of the TDA5140A chip, the disk rotates constantly at the same frequency, without stops, brakes, etc.?

How do you set the frequency of 50Hz?

Greetings,
sniff

So it's like I said: an engine with original electronics may not work, and even if it works, the vibrations are huge.

The driver is TDA5140A (there were a few problems with this typo ). This driver is friendly to use, works very steadily. Sometimes when a too-powerful engine (eg from some RC model) connects to it, it heats up quickly and turns on the thermal protection, but in the case of motors with a hard drive, this is not likely to happen. I did not set 50Hz frequencies. Simply put, I used the app shown in the catalog sheet and it worked. The rotational speed does not have to be constant - the program has been written in such a way that the image adapts itself to the current rotation frequency (it must only be between 35Hz and probably 60Hz). My engine has 3 pins, but four motors are also suitable.

Hello
As I have done (thanks to the author's help for what I am THANK YOU once again) I will add a few of my own observations to this copy of the project:

- Vibrations can be caused by both too high RPM (drive with the original HDD driver) as well as the unbalanced disc after cutting - you glued something under the plate to balance it?

- It often happens that the turnover on the original application is too low - the clock does not work - I had it in 3 cases with 5V power supply. As the changes in the control capacitors did not bring the expected result, I used a less professional solution - I led the controller itself out and fed it with 6V and watch through 2 diodes (not with the Schottky barrier) from the same 6V. The revolutions have picked up and everything is very stable. The TDA logic is of course powered by the processor.

- Setting the reflection sensor gives you a bit of problems but when it hits you ... Impressions incredible. Both fields - transmitting and receiving should be minimally visible above the edge of the plate. distance between the sensor and the plate approx. 1.5 mm

- The best discs for this disc are those with a flat side surface with a thickness of approx. 1.5 mm (there are 1mm plates and a rounded side surface - I could not start the clock on this)

I hope that the insights will help.
greetings
M

Hello
I will add 2 observations from myself:
1. The vibrations can be easily eliminated by proper tightening of the screws fixing the plate.
2. To facilitate the detection of the gap by the sensor, its edges can be painted over with a black marker.

Good luck in construction

Thank you very much for the answers.

1. Yes, I glued the disc to a little improve the balance after cutting the gap.
mlassota: I read the whole thread and used your weighs on the pharmacy scale.

2. I plan to use a slightly different sensor that will detect the gap. I just need it for him so that the plate does not vibrate too much horizontally.
The sensor is in this shape: C and this C enters the plate.
TCST1202 - when there will be no slot at the entrance the port will be 0. as soon as the sensor hits the gap, the light will pass through it and I will receive the input 1. I think it will be easier to set than the reflector.

3. I ordered TDA5140A and I want it now enter into the scheme. However, I can not find the right library. Which programs did you use? (I use eagle).

4. What signal goes through the PA0 - MOTON port to this disk driver?

5. What is the power of your watch? Voltage - if TDA will not do any tricks - 5V, and a current of 2A should be enough?

sniff wrote:

Thank you very much for the answers.

4. What signal goes through the PA0 - MOTON port to this disk driver?

5. What is the power of your watch? Voltage - if TDA will not do any tricks - 5V, and a current of 2A should be enough?

Hello

With PA0 a + 5V signal goes (The program checks the code from the remote control and changes it if it is an on / off signal that changes the status of the A0 terminal from 0 to 5V or from 5V to 0 - this is the power supply of TDA logic)

2A 5V is enough with the supply. When the white face takes 1.2A (my copy0

As for the sensor ... all the fun is that it can not be seen.
greetings

Is there a way to avoid triangular hints? For example, a gap in the opposite triangle? What if I used a disk with 2 tokens and under each led? And then the minute hand could have a different cut

Unfortunately, the triangular hints will be more or less visible (this is due to the fact that the place closer to the middle of the turn overcomes the shorter path, despite the same speed). A gap in the form of an "inverted" triangle should improve this (but may worsen the edge's sharpness).

As for the two lengths of instructions - you only need to cut the second plate (the one underneath) (in addition to the notch identical to the plate on the top) "from the inside", eg at an angle of 90 degrees and slightly rework the program.

1. How to connect the motor with 4 lead out under the TDA5140 ??
2. Is it possible to obtain RPM regulation using TDA ??

mateo238 wrote:

1. And how to connect the motor with 4 lead out under the TDA5140 ??
2. Is it possible to obtain RPM regulation using the TDA system?

As for the first point, it can be done in two ways.

The first is to connect the three motor outputs as shown in the diagram, and the 4 pin (center of the star) you leave unconnected.

The second way is to connect three pins to the legs MOT1, MOT2 MOT3, the center of the star to MOT0 and remove the resistors 10k, which connect the inputs MOT1-MOT3 with MOT0 (exactly as in PDF from TDA5140 on page 8).

What should be the power supply on the Vp and Vmot line ?? Because it gives you 5v (from a computer power supply) and somehow it's slow to everything

It should be "appropriate", ie:
Vp - 5V - is logic power supply - does not affect the engine speed.
Vmot - this voltage has an influence (just like another parameter - the value of 2 capacitors) per revolution. As my disk also turned slowly, I pulled out the BLDC driver outside and fed it in the following way:
Powering the "main" watch 9V, 1.5A switching power supply
Power electronics of the watch - 7805 + filtration
BLDC power supply - lm317 with selected resistors at 6.6V - at this voltage the rotation speed of the disk is appropriate to obtain a stable image of the clock.
Vp pulled out of the disk plate and attached to the 8 foot of the external BLCD.
Before I even pulled the BLDC out of the housing, I did tests with the change of capacitors at 9 and 10 TDA pins, but it did not bring the expected result, hence - maybe not very "nice" but certainly effective solution
I think I have described it quite clearly
greetings
M

I understand that the capacitors at pins 9 and 10 have an effect on the speed of rotation. In the catalog note there are 18nF capacitors. I put on 20nF at home (I've had one). To what extent can I modify these capacitors (how much increase or decrease the capacity) ?? Will I be able to regulate motor speed smoothly if I make a voltage regulation system for Vmot in the 5-15V range? At what voltage on Vmot do you get good results?

Hello
I did tests with capacitors within 1-22n - the smaller the higher the turnover. Unfortunately, the regulation at 5V was too small. By adjusting the voltage, you can smoothly regulate the speed of rotation. I got good levels of rotation at 6 - 7V. When loading the disc, the motor takes about 150mA.

I have a question, the TSOP348 diode, how many kHz must it have?

It is not a diode but an integrated IR receiver and must be on 36kHz

Can TSOP4836 be mounted instead of the TSOP34836 IR receiver?

Rather yes - I also did not use exactly the same receiver - pinology is correct

Sh44dow wrote:

How old was your drive? What capacity?
My 40-gig IBM's plate crumbled like a glass when it touched the corner It also crumbles when cut with a handball.
I found an old plate from a dozen-year-old disk ~ 120MB (still with the "golden" layer from the outside) and cut out without any problem in it. Unfortunately, it is crooked like a sidewalk in the city and is not suitable. .
I have a few more 9GB SCSI disks that I used to buy for a penny and I'm thinking about undressing one I would like to avoid unnecessary demolition so I need to determine if only newer disks are so fragile .

Disc disks are made mainly of glass or aluminum alloys. If there is glass under the magnetic coating, it will fall apart from impact or other mechanical treatment, but if there is an aluminum alloy under the magnetic coating, it can be safely machined.

Added after 6 [minutes]: < br />

 

Press wrote:

Nice idea and great performance.
But that's how I wonder how to make tips different in length ...
Anyone have an idea?

Maybe he would use something like this http://pl.wikipedia.org/wiki/Nadbery



Added after 5 [minutes]: < br />

 P.S .:
In the German Wikipedia Nip shield is better described and shown on a Polish or English wiki - http://de.wikipedia.org/wiki/Nipkow-Scheibe

I am asking for help. I did everything I sold and cramp somewhere short circuit because it disconnects the programmer; / and I can not even program it, someone has an idea where it can to lie?

Hello
It's hard to deduce anything from these photos.
Maybe the jet processor shifted on the pads? The short-circuit is quite easy to do both at the "top" soldering of the cabinet as well as under the programmer's socket.
You have to carefully follow the flow of power - it is not complicated in this project. I have given three critical points. Another critical point after looking at the plate is the 10uF capacitor - the bottom right corner of the plate - the pads are under tantalum - so they are quite close - it can move - for now it can be soldered out - the watch will also move without it ... In the myriad of possibilities straighten the parts they lay exactly on the pads.

The system now programmed a question how to connect the TDA to an engine that has 4 pins? How do I know which cable from the HDD motor is the center of the star? And can you power your watch with 12V?

Hello
Write what was wrong - for others
Take a universal meter. Set it to the resistance measurement with the lowest value.
Only between one of the four ends and the remaining resistance will be the lowest - it's the center of the star and you leave it alone.
You connect to the other three. the motor should turn to the right, but if it rotates to the left, you change the order of any two ends on the engine - it will start to rotate in the opposite direction.
I hope I explained it quite precisely.
greetings
M

I have a meter only it is broken has a small resistance; / shows non-stop 2 Ohms, there was a little solder paste at Atmega and now worse problem ; / by the question I connected the voltage back; / the atmegia started to smoke; p later, as I have already figured, I connected normally and the patrol burned out a size hole like SMD 405 and started to shine. lol and you need to buy a new; / could you go TDA too? or something else?