Maintenance

[Penri]

Terry

Are the timings broadly in line with your expectations, given my manual timing inaccuracies?

Penri

[TerryJC]

Penri,

It's difficult to tell, given that there should be two loops in each segment. I would like to know if the sensors are detected (even though they probably haven't been passed), because that will cause an immediate deceleration which is specified as 2 seconds.

It should be:

• Accelerate to max speed (approx 2 seconds).
• Run until the sensor is detected.
• Decelerate to min speed (approx 2 seconds).

On the face of it your timings may well be OK because there are 3-4 seconds of running time between reaching max and starting to stop. I don't recall if that was the original timing in 2017.

[Penri]

OK I understand.

[TerryJC]

• Accelerate to max speed (approx 2 seconds).
• Run until the sensor is detected.
• Decelerate to min speed (approx 2 seconds).

Penri,

Something was niggling me; where is the delay between reaching full speed and giving up looking for the sensor? The line above 'Run until the sensor is detected' should be followed by 'or timeout if sensor not detected after a specified time'. In the code, this defined as 7 seconds and the code tests for the sensor crossing every 0.25 seconds, so the total run time should be a bit longer than you are seeing,

That implies that the sensor is not being reached, but is being detected anyway. That's difficult to diagnose without monitoring the running code (there is a printout "Sensor B Activated", which should occur sometime before the 7 seconds is up on the out leg. This code and its associated delays were fine tuned in situ to give a fighting chance for the sensors to be detected but also prevent the train going too far (the original 'buffer overrun' ).

It looks as if the code is actually detecting the event too early. This was tested on the Board before I brought it back, but it is quite difficult to simulate because I have to press a button at just the right time to generate the sensor passed signal. It is possible therefore that I saw the false trigger and thought it was the real one. If the board is working, then the sensor or wiring may be at fault a short to 0 V.

[Geoff]

The slow running could be caused by the motor failing in the drive unit or too much mechanical resistance in the pickups for the lighting but as I said it was running OK when I left Monday.
The centre rail and the outside rail carry 12v DC which is from a separate power supply (off hand I cannot say which is + or -) which should be totally indipendent from the running rails just for the lighting in the coaches.
To prevent the train hitting the buffers apart from the sensors there is a diode fitted near the end of the track to prevent that happening.

Geoff

[Penri]

Hello

Popped in today on another mission and found the underground train at the right hand end of its travel.

I wanted to see how it would fare without its coaches so I slid it back up the track, uncoupled it, only to find that it wouldn't move at all. I suspect the fuse has blown.

BTW the signal light wasn't on either, neither red or green.

I'll pop down to Halfords later to buy a slack hand full, probably 2A if available.


Hwyl

Penri

[Penri]

OK, changed the fuse, this time fitting a 2A version, and we have motion again.

Unfortunately the power car exhibits the same characteristic as the complete train, ie it shuffles it way towards the right hand end of the track, the return journey never taking it back to it's start position.

My eyes may be playing tricks on me but I'm sure the train reaches a higher speed travelling left to right rather than the other was around, which would explain why it shuffles it way to the right hand end. Why that should be needs some deeper thought.

I bought the "dud" fuse home to test and lo and behold it conducts, strange.

Penri

[TerryJC]

Penri,

Do you get the sense that the train goes slower than it used to?

[Penri]

Hummm!

If pushed I would probably say yes, it certainly (all else being equal, which they are probably not) does not get to where it should be after each acceleration - deceleration event.

Penri

[Penri]

Good morning

Spent a little time this morning capturing some waveforms from the underground railways tracks. The scope probes were hooked directly on the tracks:
Yellow trace- Middle rail
Blue trace - Outer rail

IMG_7028.jpeg (132.36 KiB) Viewed 301 times

The vertical scale of the scope display is 10 times the actual measured value.

No motion (TEK0000)

TEK0000.JPG (235.39 KiB) Viewed 301 times

Left to right (TEK0001)

TEK0001.JPG (226.02 KiB) Viewed 301 times

Right to left (TEK0002)

TEK0002.JPG (248.05 KiB) Viewed 301 times

Left to right (TEK0003)

TEK0003.JPG (212.54 KiB) Viewed 301 times

I also took 2 videos of the entire journey cycle so you can see the traces changing during each section plus a video of the train emerging and returning at the right had end of the layout, after it has shuffled it’s way to that end. I only thought of using the video capture while I was at WMT so didn’t have a tripod to hold the camera steady. I’ll post these on Google Drive and send you an invitation to view Terry.

4 more traces to come.

I'm just sharing these now I haven't analysed them myself yet.


Penri