2014. április 30., szerda

FRDM-KE06Z de-bricking

On Monday morning I went to the office. There I connected the board to a Windows 7 machine. The status went to Ready from Programming imediatelly. In this way I was able to ugrade the bootloader and download the OpenSDA debugging firmware. The new bootloader (v1.11) stated imediatelly that the old one (v1.10) not compatible with the Windows 8.1. I was surprised on this a little bit. The problem is not new (http://pakahuszar.blogspot.hu/2013/12/windows-81-mired-freedom-2.html), but the board itself is a 2014 model. So I don't understand why Freescale wasn't able to put the corrected firmware into it.

2014. április 29., kedd

ARM Course 11.

At least with the evaluated things. Just the module 15 left. It would be great to finish the game until the submission deadline (7th May).
Here is the result:

2014. április 27., vasárnap

The week of the catastrophy

Since the Friday of the previous week I haven't wrote anything. This more than a week wasn't passed without events. I can tell it was catastrophic in all aspects. I just hope, that my actions will have better results on the next week. Without the demand for the correct timeline I write down, what happened:

I'm at home for a week now. I'm sick most of the time I was lying in the bed, reading. Tried to cure my ear/eye/nose/throat etc. problems. My left ear is still not working.

I was trying to build the CAM software named HeeksCNC. First I tried on Windows. It won't work. The compiler gave some insane error.
After this I tried it on the fresh Ubuntu 14.04 LTS. After more than two days struggling finally worked. The build and not the program. When I started the application and tried to create a new tool in it, I got a segmentation fault. I could spend additional days with it (most probably the versions of the OpenCascade/OCE is the problem), but my goal is not building but using it.
So I installed the trial of the Windows based official build. This at least not freezing just time to time stops for a few minutes then goes forward. The problem not solved, and will not to be for a while.

I ordered two of this board. Earlier I bought a Pololu Zumo robot for my son, and I planed to assemble it sometime. The robot itself is in one piece for a time already, just I didn't wanted to use put an Arduino on top of it.

Instead of the Arduino I wanted to use a KL25 based freedom board based on Erich Steyger's work. Finally I didn't liked the problems with the 3.3V/5V power supply so I put it aside. Now this board seems to be the solution because it works on 5V.

I bought two, to be able to experiment with it after finishing the robot.
The first mistake I made that I soldered not just the Arduino headers in but also the header at end of the board. This mechanically collide into the Zumo blade. No problem, this board will be my experimenting one. Picked the second board and soldered in without the header in question.
A few days later I started to play with it. I tried to put a sample application to it. Since that just the bootloader pops up, the LED blinking and according to the status file it is in "Programming..." state. Forever.
I managed a brick from it. After a few hours of searching I was able to find out nothing (to be precise I seen a comment somewhere that the Windows 7/8 may can correct it, but I've only 8.1 at home already).
The actual status: I need a new board.

3D Printer
To exchange the printhead to the new one I have to print one additional component. The new fan holder. I'm trying to achieve it for the third time (I wrote the whole story already just didn't published it, on the next week I'll takte the time to finish the part of my half finished articles). I cleaned the printer and started the print.
It doesn't start.
The heating of the build platform switched off around 85°C, and I wasn't able to switch it back. A few days later I retried it. The result is the same (at least the heating started and not completely dead).
I remembered from somewhere, that others also had similar problem with the heating of the build platform. They put an SSR into the control circuit.
It should not cause problem. I try it.
With half hour hard work removed the pőower supply from it's place. I've more and more the thought the I would never honor them an "Excelence of engineering" kind of award.
I put in an SSR removed from an old equipment.
The heating won't start.
Let's measure.
I set my DMM to 20V. Measured the input of the SSR. The  instrument signals overvoltage. Set to 200V, the result: 178V. Switched off  everything and removed the load's cables. The PSU without load: 178V.
Thinking, cooling down. If it would be true the circuits would already produce smoke and smell. Switch the instrument to 1000V, the measured value: 12V. I thing I had a DMM.
Picked my old, half dead Metex out of the drawer. It measures 12V everywhere.
Reconnected everything, put a new SSR intead of the savaged one.

The heating started. The SSR also heats like hell. The 25A reated SSR will not be enough (and of course I didn't have a bigger at home). For a time being I put a heatsink under it, to see the outcome.
It heated the platform correctly, but it can't be used in this way. I've to order a bigger SSR. This is waste of time until it arrives.

DMM - continuation
The end of the story that I don't have a working DMM (I should build one. :-D ).
My Metex is 25 years old. It has permission to give up. To be precise it is not dead just the resistance measurement not realy works (I think that is the rotary switch).
The other one the Maxwell - it looks like a Chinese crap to me (Don't misunderstand me! I've no problems with the Chinese goods. There are manufacturing good and bad things like everywhere on the earth. This thing especially not on the good side). But because of it's capabilities I payed more than €50 for it a few years ago.
It is enough to me. The Metex doesn't exists anymore (or at least I don't know where to find it). Except of the big names (Fluke, Agilent) the things what man can buy (Maxwell, Tenma, Voltcraft by Conrad, etc.) I've no idea which is "instrument" and which is crap.
So I ordered a basic Agilent DMM.

This even costs €110 to me but with 2-3 trial I can spend same amount of money to crap and the outcome would be questionable.
The good thing on it that I'll get the replacement of the bricked FRDM board with one shipment cost.

And on the positive side
It is just a sidenote: I reserved some time to the ARM course and finished three modules (11-13) and all of the quiz. So just the 14th lab and the 15th module (what is not evaluated), the creation of the game project left behind. It not seems impossible until 7th May.

2014. április 18., péntek

Ubuntu 14.04 LTS

Yes you see it correctly, this is an Ubuntu. To be precise an Xubuntu Desktop 14.04 LTS.
It was released yesterday. I installed it today dawn.
Yes, I'm a kind of  Windows guy. And?
There are some things I need nowadays and not available for Windows. So from now Linux also.

2014. április 10., csütörtök


I'd like to give some attention to a crowdfunding campaign. Actually a Hungarian one. I've three kids. They are cycling.
Here is the product:
I think it needs no more explanation.
The campaigns link is on the right panel.

2014. április 7., hétfő

Robo3D - new Merlin firmware

Yes, your eyes not cheat on you, this is an Arduino IDE
How deep I went. No, no, I'm not writing code in this. It was needed for the firmware upgrade of my 3D printer.

3D things - Thingiverse

I'm working more and more on 3D things. I started to design things what maybe useful for the public. Therefore I created a Thingiverse account to publish theses (there will also be available on the GitHub).
Location for my Thingiverse stuff if something finished:

ARM Course 10. - FSM

After a long skip continued the ARM Course. The section 10 is the hardest part according to the instructors. My opinion is the same. And it is the most useful. I've gaps in my knowledge for shure, or maybe my education was incomplete (I was learning embedded systems something like 20 years ago in the college), but this is the first time I've heard about FSM (Finite State Machine). It a very simple simple method for subscribing and programming something what is required for an MCU based system on daily bases. I'm absolutely sure that I'll use it frequently.
I finished both the quiz and the lab.
Her is the working traffic light: