SmartiPi Touch from SmartiCase

I have treated one of my Raspberry Pis to a new 7-inch touchscreen which I intend to house in a SmartiPi Touch from SmartiCase. This is another project I have sponsored through Kickstarter.

A freshly prepared SmartiPi Touch.


The corresponding 7-inch touchscreen ready for mounting.


Unpacking and checking the parts.


As you can see I have ordered the camera case for later experimentation. The allen key is supplied to fit the hinge bolts as well as an option to discard the base and mount directly to a larger surface or even a VESA monitor arm.

In the next post I will describe how to assemble the screen in to the original base. In the meantime, why not go to the SmartiCase website and have a look…

Marconi Signal Generator Part 3 – Repairs and Testing

David 2E0DYN and myself spent two days in the lab working on the signal generator, removing old capacitors, cleaning here and there and finally fitting new components.

One problem we encountered was the poor state of the original mains filter capacitors. The plastic bodies had either cracked and broken, or the ends of the bodies were coming away from the wire. These would all be replaced.

The mains filter caps were in a poor state
The original devices were replaced with suitable safety rated devices
The original devices were replaced with suitable safety rated devices

In many cases we had to replace old ‘can’ style electrolytic capacitors that would have dried out over time. These would be replaced with modern ‘motor’ capacitors that have similar high voltage and capacitance characteristics.

‘Old’ and ‘new’ style electrolytics side by side

The old ‘can’ in the above picture would pose a challenge, as it contained multiple capacitors in one device. We would have to find some way of replacing it with a bank of individual capacitors.

‘Hybrid’ capacitor replacing original multi-device ‘can’

There were also some alterations made underneath the power supply. The original capacitors had large lugs on them that allowed other wires and components to be directly soldered to them. The new ‘motor’ capacitors had a pair of flying leads instead. It was necessary to fit some tag strips to allow everything to be reconnected.

Tag strip allowing connections to be made
Tag strip allowing connections to be made

Finally the power supply section was ready, having had the necessary component changes and tag strips fitted.

Underside of completed PSU section

There were of course many other components to replace, including several that required some major disassembly of the front panel to allow access.

There were some hard to get to places needing lots of work with screwdrivers and small spanners

Things were progressing well by now. The replacement components were all fitted and work began on the physical reassembly of the unit.

The unit is reassembled prior to testing
The unit is reassembled prior to testing

Finally after a few hours of the second session in the lab, we could do a test power-up of the device. This would be the first time in several decades that this unit had come alive.

The device is working for the first time in many years
The device is working for the first time in many years

Here is the output on the oscilloscope. (Yes, the Gould scope is back in the equipment stack. The HP is fun to use but it is very noisy). A nice clean waveform was observed. The set was then hooked up to my DX394 receiver where we were able to test the output of both continuous wave and amplitude modulated signals from the unit.

A clean CW waveform
A clean CW waveform

Marconi Signal Generator Part 2 – Assessing what needs to be done

I knew that when the signal generator arrived on the bench, it would not be safe to just plug it in and ‘suck it and see’. There was a very good chance that some components had failed with old age and this could result in some fireworks and a severely damaged unit if power was applied.

Unfortunately I have little experience in restoring old valve-based equipment. However, I was able to call on the expert advice of David 2E0DYN who is quite the boffin on valve equipment in general and even better, he has recently renovated his own TF995A. My unit was in safe hands!

We set to work, removing the six bolts in the front face which allowed us to slide out the main chassis from the case. The power supply is built in to the bottom of the case and is connected to the chassis by a six-pin plug.

The chassis connects to the PSU with this six pin connector plug

One interesting aspect of this renovation is that unlike a lot of modern equipment, the TF995A is clearly designed to be disassembled for servicing and repair. We had access to the original Marconi instruction/service manual which allowed us to compile a list of components that would need replacement.

Capacitor Cans
Typical old electrolytic capacitor cans scheduled for replacement with modern parts

A decision was made at the outset – old capacitor cans such as these would be replaced with modern parts. This would not be a ‘historical’ restoration trying to return it to some sort of factory finish, but a repair with the intention of putting this unit back in to service to do useful work and add more to its life story.

Capacitor Bank
Two banks of capacitors that all need to be replaced

David’s eagle eyes spotted some failed capacitors deep within the mains filter requiring the disassembly of the filter.

Mains Input Filter
The mains input filter before disassembly

There was also work to be done underneath the power supply section. This was removed from the case and inspected for components needing replacement, as well as planning the routing of replacement cabling.

PSU underside
The underside of the power supply showing some of the wiring

Finally, it would be necessary to strip down part of the front panel. This would permit the removal of a sub-assembly attached to the back face of the panel. Sandwiched in between this sub-assembly and panel were more components needing attention.

Stripped Panel
The front panel was stripped down to allow access to screws holding the sub-assembly in place

After a few hours of work, we had a list of components to buy and jobs to be done. The next post will show the work being done.

Marconi TF995A/5 Signal Generator

I have an uncle who for for many years up until his retirement worked as an engineer for the UK Civil Aviation Authority. At some point during his career he was able to purchase an old Marconi signal generator that has become surplus to requirements.

Years passed and the signal generator quietly gathered dust up in the attic until a recent clearout saw it delivered to the workbench of G7IVF to see if possibly, just possibly,  it might find a home here. Silly question really…

However, because the unit hasn’t been powered up for a considerable amount of time, it is safest to assume that some investigation and repair work will be necessary. More to come on this…

Meanwhile here are some pictures of the generator as well as an interesting piece of history in the form of a label that was attached to the unit showing it was withdrawn from service and placed in to storage in 1976 before being sold on.

Marconi signal generator image 1
The Marconi TF995A/5 sits hopefully on the bench of G7IVF
Marconi signal generator image 2
Bright sunlight after so many years in the attic. The Marconi is suddenly full of hope…
Marconi signal generator image 3
Yes! It’s going to become part of the radio station G7IVF. A happy signal generator!
Marconi label front
Part of the ongoing history of this unit – the original label from the CAA.
Marconi label rear
The last time this unit was used properly – 1976. Forty years on, the story is about to continue…

Preparing and Fitting the Edelux II Headlamp

Having unboxed and inspected the new lamp it is time to prepare it for fitting to Rose, my trusty Thorn touring bike. First, I used the old wiring harness from the original headlamp as a template to cut the new cable to size:

The new lamp cable cut to the same length as the old one.

Next, I stripped away a few centimetres of the cable sheathing to expose the two inner conductors:

The conductors after the cable sheathing has been stripped away.

Because these two conductors will be close to the road surface at the centre of the wheel hub they will need protecting with some heatshrink tubing that is supplied with the new headlamp. I use an electric heat gun to do the shrinking. It is like a hairdryer… but much much hotter!

The hot air gun ready for action.

After cutting the heatshrink tubing to the correct length, make sure to use a suitable tool for holding the work while heat is applied. Remember, the air from the gun is very hot and it only takes a few seconds to complete the task.

The first heatshrink piece cut and ready for heating.

Here is how it looks after shrinking down:

The protected conductor cooling down.

Once cooled, another piece of heatshrink tubing is applied at the point where the two conductors enter the cable sheathing.

Extra protection for the conductors.

Now the conductors are ready to have the crimp connectors fitted. It is easy to fit them with a small pair of long-nose pliers.

The crimp terminals fitted ready to attach to the hub generator.

As you can see, the crimps each have two sets of tabs. One set of tabs is bent over the copper conductor to create the electrical connection. The second set of tabs wrap around the insulation to add mechanical strength to the connection.

Preparing the crimp connectors to survive the open road.

Time to add more heatshrink tubing, this time to protect the crimp connectors from damage.

Ready for the road!

The electrical connections are now prepared. We can now fit the headlamp to the bike…

Rose has her new headlamp.

The new headlamp has been fitted. As you can see, I had to replace the mounting bracket as well as the headlamp.

Wired up and ready to power the headlamp.

The crimp connectors take their place on the hub generator ready to supply power when needed.

The cable is secure against the front fork.

The cable runs up the front fork as before and is held securely in place to prevent it fouling the brakes.

Ready once more for the daily commute.

Rose is now ready for the next commute although it looks like she could do with a bath (the riverside path that forms part of my commute has been quite muddy recently). You can also see the rechargeable headlamp on the top of the handlebar that provides extra lighting and some redundancy if one lighting system should fail (I also run two tail lamps for the same reason).

Rose has a broken headlamp

Poor Rose. Years of cycle commuting has taken its toll and her hub-powered headlamp has fallen to pieces.

Disaster strikes! Rose has lost her headlamp.

I have chosen to replace the headlamp with a Schmidt Edelux II LED light from St John’s Street Cycles.

The new lamp has arrived.

I like the simplicity of packaging. It suggests that all the effort has gone in to designing the light, not the box it comes in.

The box contents at first glance.

The new headlamp is unveiled. What else has come with it?

Everything is supplied to complete the replacement.

That’s everything needed to complete the replacement. The extra coiled cable is for running to the rear of the bike to power a tail light. However, Rose has two large Cateye LED lamps powered by batteries that help keep us visible at night.

As you can see, the lamp supplied is the version with the 140cm cable which will need to be cut to length and prepared for fitting to the front fork. More on this in another post…


Changing my Oscilloscope

I have a few days off from work and it seemed like an ideal opportunity to tidy up my electronics bench. This would allow me to change out my old Gould 20MHz oscilloscope for something different.

In this case, the something different is a Hewlett Packard 54200A Digitizing Oscilloscope. I know it powers up, but as yet I don’t know if will serve as a reliable instrument for my experiments. Time will tell…

HP 54200A
The new oscilloscope installed on the workbench
The tidied and reconfigured workbench

Upgrades to my Pi fleet

I have recently expanded my collection of the little Raspberry Pi computers as MyGoodLadyWife has bought me a Pi2. I am hoping to be able to use the extra processing power due to the increased clock speed and quad-core CPU to good effect as a controller for a software defined radio. (More on that to come…)

But first, I needed to install an operating system. I did not want to simply transfer the existing SD card from the Pi B+ that Hector was proudly sitting atop of in my previous post. This is because the Raspbian image on that card is crafted to work on an ARMv6 CPU, whereas I can now use an image for an ARMv7 ‘HardFloat’.

I wanted to stay with Debian on my Pi2 so a quick trip to the relevant page on the Debian Wiki shows that there is a suitable image available for Debian 8, otherwise known as ‘Jessie’. So, I opened up the SpartiPi case and swapped out the old B+ for the new Pi2 (note that the case takes the Pi2 with as much ease as the original B+ it was designed for.)

So, all set to introduce the Pi2 to Debian Jessie. What steps are needed to make it all happen?

First, follow the link from the Debain Wiki to ” Debian Jessie on Raspberry Pi 2 by Sjoerd Simons of Collabora”. This will take you to a page with more background information as well as a link to where the bootable images can be found. I ended up downloading the following files:


Great! I have the image file, but what does that bmap file do? On his page, Sjoerd recommends using a tool called ‘bmap-tools’ to write the image to the SD card. I had to pause to check if it was possible to install it on the machine I use to program my SD cards and yes, it is. There is a repository for OpenSUSE 13.1 which I currently use (as MGLW is very fond of Geeko the Chameleon, the OpenSUSE mascot!). I added a repository to Yast which I named ‘Tizen’ with the following URL:

This allowed me to install bmap-tools after which I was able to bring up a root console, change to the directory where the image and bmap file were stored and use the following command to write the image to the SD card:

 bmaptool copy jessie-rpi2-20150202.img.gz /dev/sdd

While at first glance it seems a lot of effort to install and use this tool it does have an advantage over using dd in that it gives an indication of progress in terms of percentage completed and it also gives an indication at the end that the process was successful.

The Pi2 is now on the bench, powered up and running Debian Jessie for ARMv7. I have been able to use the apt-get update and apt-get upgrade commands successfully so it looks like it is working correctly.

More to follow on my task of putting this little computer to work in my radio shack…

Introducing the SmartiPi

Just before last Christmas I took part in a kickstarter campaign to help fund the SmartiPi case, a Lego compatible case for the Raspberry Pi B+. This is what one looks like after it arrives in the post.

Look what turned up in the post!

These are the parts contained in the package – The case, a Lego compatible top-plate, a camera case and a GoPro camera mount. (Oh… and some screws).

The contents of the bag, at large on the G7IVF bench.

The Raspberry Pi slots in to the case without difficulty.

The Pi fits easily in to the case.
Another view of the Pi B+ in its new home…


Four screws close the case, and it is ready for powering up!

Ready to be powered up.
Ready to be powered up.

Make sure you have plenty of extra Lego to hand!

Hi there! Grab a brick and join in!


Update on the 525

The NRD-525 that was recently on the bench has gone back to the M6BOH/M6JOJ shack, but not before I had a chance to use it on my recently constructed random long-wire and 9:1 un-un. I had some fun getting the hang of decoding HF weather faxes using the FLDigi package which can be obtained from W1HKJ Software – home of FLDigi

Typical HF weather fax image received using JRC NRD-525
Typical HF weather fax image received using JRC NRD-525

The image looks slanted because I have not spent time calibrating the sound card on my laptop. I’ll come back to that another time…