Fusion Electronics Library Notes and Crib Sheet

After spending many hours going round in circles trying to create new custom library parts in Fusion Electronics (FE) I gave in and posted a plea for help on the FE forum. I received some helpful replies but not sufficiently uplifting to put me at ease with the process.

As ever my nerdy side stepped up and armed with this new knowledge I set about learning the process step by step in a way that I could understand it and more importantly repeat it successfully.

The result of this is a 30 page document that can be downloaded from the link below. This contains the library process, a help crib sheet for using FE and the copies of the original support replies I received from the FE forum.

It may not be perfect and it is a work in progress so feel free to give me feedback on errors and content changes. Remember it is based on Fusion Electronics as of today at version 2602.0.71

I hope it helps someone, somewhere, sleep better. The length of the waffle will almost certain guarantee the latter.

Links to similar or related post are listed below : –

CCTV System with remote camera over WiFi

I had a request for help with a CCTV system where the owner wanted to add a remote camera to an outbuilding with no cable route possible. This should have been a JSN job (Just Say No) but I’m always up for a (steep learning curve) challenge. After some investigation (back of a napkin sketch and a quick Google search) I opted for a 5GHz WiFi bridge and thought the job was going to be easy. JSN jobs are like that but you have to be conscious that they come configured with inbuilt mission creep.

The WiFi link was a 24V power over ethernet (PoE) system. The existing three cameras were 48V PoE and the additional camera was to be the same.

Here is the final configuration. The red lines are PoE and the blue lines are just basic CAT5 unpowered connections.

The NVR is a SPRO four port device with 48V PoE on all ports. To save cabling the owner had used a PoE pass through switch so two cameras could share the same cable running back to the NVR. The existing third camera was a direct feed from the NVR. I was quite impressed with the pass through device. It has the capacity to allow four cameras to share the same PoE NVR connection.

Back to the plot … Adding the fourth remote camera using a bridge would have been simple if the bridge was 48V PoE powered but it was 24V PoE. The remote location therefore needed two PoE inserters – a 24V one for the bridge and 48V for the camera. The LAN ports on the two inserters just need connecting together to complete the link. The NVR ‘sees’ the remote location as a locally connected camera.

Something to note. The bridge (KuWfi) as delivered has two IP addresses configured, 10.1.1.253 for the Master location and 10.1.1.100 for the Slave location. All the cameras are allocated 10.1.1.x IP addresses on the NVR. The NVR also has an internet connection for remote Smartphone viewing.

KuWfi provide an application to configure the two bridge units. I spent a long time trying to work out how to get this to respond to either device. In the end a light bulb moment reminded me that a previous Chinese sourced product had the same issues. I had traced this to a Chrome related problem (which I never solved) and found the Application would only work using Edge. The same happened this time. Using Edge got me into both units without a problem. Note that you have to set your host interrogating computer to be on the 10.1.1.x domain.

There is a signal strength graphic on the Slave GUI which is very useful. You can connect your computer LAN port to either the Master or Slave 24V injector LAN port and request either of the two addresses (10.1.1.100 or 10.1.1.253) from either end of the link

Just think how much simpler it would have been if the bridge also ran from 48V …. maybe someone produces a 48V to 24V PoE step down converter ? Or maybe I should just stick to the things that I pretend to know more about … and JSN.

Links to similar or related post are listed below : –

Qidi Slicer auto support error on my part

The other day I modelled an adapter boss that had a threaded end to fit a standard CCTV camera mounting box (GB Pipe Thread – 26.441mm x 1.84mm tpi) and with the other end to suit a 20mm flexible conduit. Here is a Fusion pictorial view.

I modified the thread profile as per my post on thread tweaks and posted the file to Qidi Slicer. I opted to print in the orientation as shown above. This clearly needed support under the shoulder so without thinking I ticked the auto support tick box and hit print. The thread was a disaster flecks of filament such that it did not mate with the camera mounting box threaded hole.

Head scratching time, what had gone wrong ?

I had thoughtlessly just ticked autogenerate support and Qidi Slicer saw the thread profiles as potentially needing support and hence all the flash filament. Re-running with ‘support on build plate only’ removed the problem. See below.

This problem probably become more apparent due to the large size of the thread being modelled. Probably a dumb thing to have done but I tend to learn more from doing dumb things. One issue to be aware of in the future.

Links to similar or related post are listed below : –

Error Code A9 on Odealis Gas Boiler

Not something of general interest – our house in France has an Odealis LPG fired boiler. This provides both hot water and central heating and works very well and is good on efficiency. The hot water side is particularly good in that it is on demand when you run a hot tap and has a limited buffer volume of hot water stored internally.

For sometime we were having problems which resulted in an A9 error message being shown on the control panel. We had a service engineer call twice on this issue and no real resolution. The symptoms were that on demand for hot water the boiler would fire up, run for a short while and then the over temperature cut out would kick in and the boiler stop firing until it had cooled down sufficiently.

If you look in the handbook for error code A9, it tells you to check the temperature measuring thermocouple. This checked out okay and was not the issue.

The boiler has a device called a plate heat exchanger (French term). It is a laminated stack of copper capillaries that have both the hot water and the central heating water flowing independently through them but thermally coupled. It is a heavy lump of metal.

My thinking was along the lines of if this gets blocked the flow reduces through the capillaries and the water temperature rises as it cannot be easily conducted away. I couldn’t decide which side could be causing the problem. We do have a water softener installed which would tend to suggest that the problem was not on the hot water side but more likely on the central heating section.

I requested a service call and this time told the supplier unequivically that I wanted the heat exchanger replaced and that he should leave the old unit in the utility room for me to inspect on our next visit.

On arrival the over temperature problem had gone away. On inspection the old heat exchanger was clogged with metal debris liberated from the inside of the radiators. The capillaries are so fine that this was always likely to become an issue with age and usage.

Having replaced the heat exchanger, I decided to try to alleviate future problems by fitting a Vortex magnetic trap in the boiler return feed. This is a large Neodymium magnet in a trap that forces all the return water to circulate around it and grabs the metalic debris. The trap can be valve isolated so the magnet can be periodically removed and cleaned.

Hopefully the problem has gone away and the new exchanger is better protected from blockage with the magnetic trap doing its job.

Sorry not very interesting but just might save someone the same frustration we had with a supplier who didn’t really understand the source of the problem.

Links to similar or related post are listed below : –

Power banks that go to sleep with low current drain

I am working on some very low current Arduino projects using the Seeed Xiao. These use LoRa radio connectivity and I needed to go walkabout with the receiver module to check the radio coverage. A power bank seemed like a useful portable power source. The problem with some power banks is that they auto switch off if they do not see 50mA or so of current drain. My receiver was using much less current than this so the power bank kept turning off. Frustrating and irritating.

The good news is that one of my power banks is a Charmast device as shown below and linked.

Thoughtfully this device has a soft workaround. If you plug your device into the USB-C connector and turn on the power bank you will see the battery capacity displayed. If you then hold the power button in for a few seconds, the display changes such that the last digit of the readout cycles round its constituant segments every 15 seconds or so and the power will remain on.

I am guessing that there is an electronic timer in the power bank that pings a surge of current every few seconds to disguise the overall low current drain.

Not all power banks are made equal.

Links to similar or related post are listed below : –

Verified by ExactMetrics