Phil's Arduino Tutorials Blog...

I've decided to keep a Blog just to add any up to date information, experiences or news but only where it's relevant. Knowing me, I'll be off-track before too long and be talking about all and sundry but, as I don't expect anyone to take much notice of my ramblings, it doesn't really matter. I am constantly adding content to this website and I receive a lot of feedback from visitors which results in changes, additions and new directions. I do listen to what is said by visitors and always welcome input.

Last updated: 19 March 2017 10:26

Well,  a new day and a lot has happened since my last update. For those of you interested in how to get a failed Windows 7 installation which is stuck in Windows Update going have a look here. The page is under construction at the moment but the information is all there.

16 September 2016:

 HOORAY! I received an e-mail this morning from my Son who sat down last evening to have a serious look at the Windows 7 update problem. He set up and delved into fresh Windows 7 installs which showed the usual symptoms of being unable to find and download Windows updates after Service Pack 1 had installed. What he found was both shocking and interesting. Bear in mind that if you research this problem on the Internet you will find 1001 solutions which various people say have worked for them but none of them worked for me.

 What my Son found was that the update agent 7.6.7600.320 was corrupting the update database so that Windows Update would continuously search for updates but would not be able to log them in the database for download and installation. He discovered that if you use KB3138612 which is a later version of Windows Update the corruption did not occur. I will add more detail on this later in case anyone is interested but I am watching my first successful Windows 7 installation in the final stages as I write.

I decided to cut the entire section on my horrible experiences with Windows 10, my move over to Linux and the move back to Windows 7 following my Son's discovery of what was causing Windows 7 clean installs to fail.

Just let me say that Windows 10, in my opinion is the most awful looking experience I have ever had with an operating system. I don't doubt its underlying credentials but it just looks awful. Add its failure to run some of my cherished older applications and the amount of Spyware built into it and you end up with an OS that is seriously suspect. My Linux expedition failed due to my need to run certain Windows applications which even WINE failed to handle.

Fake FTDI UART

13 November 2014 Fake FTDI UART modules:

 If you have an FTDI UART module and it looks like the one in the picture on the left then be aware that it is very, very likely that the FTDI chip on the PCB is a fake! Other similar modules may also contain fake FTDI chips but, the one shown is the only one I know of at this time. Over the last couple of months many owners of this type of UART have reported that following an update of the driver software supplied by FTDI, their modules have ceased to work. The symptoms of the failure is that the module fails to find a driver when it is plugged in to a USB port.

The problem has mainly been seen on Windows PCs following the installation of an updated FTDI driver by Windows Update but other platforms can also be affected, such as Linux, when the user updates their drivers. What has happened is that FTDI (Future Technology Devices International Ltd), the manufacturer of the FTDI chip, got so fed up with Chinese companies producing counterfeit copies of their devices that they put out a driver that was able to identify a counterfeit device and render it unusable.

 The fake FTDI chip is NOT a copy of the real chip, it is a fudged together device that looks like a FTDI chip and works in the same way but there the similarity ends. The FTDI chip identifies itself to the operating system and driver by the use of two parameters, the VID (Vendor ID) and PID (Product ID) and what FTDI did was to overwrite the PID parameter in the fake FTDI chip so that the driver would not recognise the chip. The real PID is "6001" and the bricked PID is "0000".There has been an uproar on the Internet about this and FTDI have back-tracked by saying that, in future their driver will only warn users that they have a fake device rather than bricking it. I have produced a "fix" to make bricked modules work again and the details can be found here.

Memory and other things:

 Have you ever run out of RAM and/or Flash?: Well I have recently! I am building a Home Monitoring/Control system which may become a project at some time and I was trying to use up my spare Atmega328 AVRs. The whole system of sensors and the Master Control Hub are using nRF24L01+ modules to communicate with maniac bug's RF24Network.

 Everything was fine until I added an SD Card reader/writer and the SD library when everything started to go wrong. Firstly, despite everything compiling correctly, as soon as the code ran on the AVR the Serial output I was using for debugging began to misbehave. In fact, the print-outs in the start-up() section kept stopping with only half the text printed out so, I added a function to print out the free SRAM on the AVR and got 7 Bytes! 7 Bytes left out of 2K Bytes was obviously the problem so, I explored using a second Atmega328 just for the SD Card using Software Serial to communicate between the AVRs but it was messy. I also realised that I was already using 26K of Flash and that any more code was going to bring me near the limit of the AVR's capabilities.

 I had to make a major decision and opted for a larger AVR and chose the 40 pin Atmega1284 which has 128K of Flash and 16K of SRAM. I bought a couple for GBP 7.00 each and with a little re-coding soon had the Master Hub unit working correctly but 40 pin chips take me back to the days when just the CPU in a computer used a 40 pin chip. Nowadays, 40 pins means a lot of I/O pins but, as I'm using I2C for the display and keypad and SPI for the nRF24L01+ and SD Card reader/writer, I've got a lot of spare pins doing nothing?

A Regulator story:

 I recently dragged some old TO5 5 Volt regulators out of my stock and by old I mean about 30 years and when I tested each one they only would output around 4.95 Volts. I checked and double checked my circuit but nothing was wrong so I tried a newer regulator and got exactly 5 Volts. I was obviously seeing the affects of aging on these old regulators which is something we're all aware of but this was the first time I had seen such a clear example of aging so, I binned them all and bought some new regulators for a fresh start. Just be aware and check any old components in your scrap box before relying on them. Of course, it could just be that the original batch of regulators were below specification in the first place?

Bad Supplier:

 I have been seriously let down by a Chinese component supplier that I have used for some considerable time. The company concerned is Deal Extreme and, up till now, I have always found their service to be excellent. Items arrived within 2 weeks and were always as described and at the right price. However, a delivery went missing that was paid for by PayPal and when I started the claims process with Deal Extreme I learned the true meaning of customer service, Chinese style!

 I had waited 4 weeks and really couldn't wait any longer but Deal Extreme kept telling me to wait and that they were investigating but, it soon became clear that they were trying to get me to wait until after the PayPal claim window had expired by giving me dates for the completion of their investigations that would remove my ability to raise a PayPal claim. At one point, near the end of the saga, they said that they were going to check with their local Post Office to see if the items had been delivered which would have been quite a feat considering that the items were low value and there was no tracking number! I finally gave them a deadline to refund my money which was the day before my last chance to raise a PayPal dispute and lo and behold, they confirmed that the package had been lost and my refund appeared. The amount involved was GBP 4.73 and the items never arrived. Needless to say, I will not be ordering from this supplier again.