Herewith please find a couple of annotated photographs of the almost completed interface box, built to house the Lady Hanham SAC.
To complete I need to measure up and manufacture the cable assemblies needed to connect to the gate valves, Ethernet and DC power, then try up the internet cable assemblies with some heat shrink and labelling.
Terry, do you think the connectors should be labelled and if so what would you like each called? If labelled I'll add the same information to the cable assemblies connecting the I/F box to the float switches and depth probes.
After our discussion yesterday, I had a look of the spec of typical 12 V solenoid valves and found that they mostly need around 0.8 A to work. Clearly the SN5407 that is used in the Sump Sensor and Control Assembly isn't going to deliver that!
However, it needs to be done and while I'm still in the throes of building the Lady Hanham S&CA, now is probably a good time to redesign things. All through the development of this board I've tried to keep the design consistent, so all Assys have the socket for an SN5407, even though I don't actually populate it except for the Sump application. The addition of this solenoid adds a third configuration and it leaves me slightly uncertain where to put the MOSFET to drive the solenoid.
The most recent versions of the drawings that I posted above show my strategy for the Lady Hanham S&CA. My thoughts were that I would keep the basic circuit the same for all areas and cater for the additional requirements of the Lady Hanham area with a mezzanine board. Ideally, therefore, I would place the new MOSFET on the mezzanine and minimise inconsistencies.
However, the mezzanine is already built and really has no room on it for the MOSFET. There aren't many components on there, but much of the real estate is taken up with Veropins to allow the distribution of the power and signals to and from the A/Ds. I could remove the socket from the main board and fit the MOSFET on there, but that is far from ideal and abandons the idea of consistency. One thought I had was to somehow plug the MOSFET into the IC socket, but it would be a bit of a kludge and I can't quite see a way forward on that.
It would probably be possible to integrate the MOSFET somewhere in the cable loom or the solenoid housing.
I need to dig out the solenoid I have and check it's specs. its been a while since I looked at it and hadn't given any though to putting it in an IP rated box to protect the terminals from the elements but given probably that's required, housing the MOSFET in it too would not be too much of a stretch.
... and by the way, as I mentioned when we talked, I need to add a connector for the solenoid to the box in any case. I'll make sure it has at least 3 pins, +12V, 0V and gate control for the MOSFET.
I'm not sure it's worth going to any effort to put the MOSFET outside the IP65 box; the purpose of my post was to ascertain your thoughts on the loss of consistency caused by it's addition to the Veroboard. (I would still need to modify the Veroboard to bring out the 3.3 V signal from the Pi (which might be in a noisy environment).)
If all else failed, I could add a Daughter Board to the Mezzanine
Progress has been slow but steady on the build of the Lady Hanham SAC; part of the problem is that the Veroboard is a bit small for the job, even with the Mezzanine board. Here is a summary of what works:
Hall Effect Probe measurements for the G1 Butts Group. (This is basically the same as any SAC).
Float Switch detection for all three Butts Groups.
Solenoid Valve switching.
Hardware Unit Test Software. (I consolidated several smaller programs and extended them to cover the additional Hall Effect Probes and Float Switches.)
Here is what has to be done:
Hall Effect Probe measurements for the G2 and G3 Butts Groups. The A/D converters are installed, the Veroboard is ready and the wiring harnesses are complete. During an early iteration of this build, I discovered a dysfunctional Pi, so I have been painstakingly adding the hardware one function at a time with a full test at each step. So far I haven't found what caused the first Pi to fail.
Not really for this Topic, but related; completion of the Software Framework to cover the additional functions. This is a work in progress with some updates completed. I intend to bounce this work off Hamish on Tuesday evening at the LUG Meeting.
I had hoped to be ready for the coming Tuesday, but that looks unlikely with other commitments.
For my part, as far as the Hanham groups are concerned, I now have a complete set of cables to connect to the float switches, probes and solenoid. Also available are 2 completed float switch assemblies and the wooden chassis for a third turned to size, need to make up the brass switch actuators and assemble that one.
I still need to create concrete boots for the probes when you can release a couple to me Terry, no rush as they only take a few minutes to make then a day for the concrete to cure.
With a bit of luck and dry weather I may get around to laying in the Cat 5e cable from the back of the railway room to the Hanham room control box location. I intend to run the cable from the Railway room to where we intend to put the control box first without making any connections. When we're ready we can then cut the existing cable, route it down to the Control box and remake the link to the railway room using the new cable.
I tested out the 12V water valve solenoid yesterday, the current draw was around 350mA according to my bench power supply and I didn't see any current limiting on switch on/off.
I now have a working Lady Hanham SAC Assy plus a fully debugged Unit Test program. In due course I will publish the updated documentation, but before that I want to commission the updated Software Framework. More about that in the appropriate Forum Topic.
I have pushed sac.py to GitLab and deleted the old files.
They were all early attempts to test the hardware. There were two called hall_effect_probe.py and hall_effect_probe2.py, plus one called level_shifter.py (I think).
