Ideas for treatment plant alarms

[Jeremy Harris]

I've started designing a remote alarm/air pump pressure system, prompted by this series of posts quoted from the closed Ebuild forum:

Quote

Sewage Treatment Plant Alarm

#1 Alphonsox

I am trying to determine whether our sewage treatment plant is required to be fitted with a failure alarm of some kind. When we purchased it a year or so ago an alarm wasn't included but I notice that most currently sold now have some system to indicate mechanical or voltage failure.

Can anyone comment on this or point me in the direction of the technical requirements.
Thanks

#2 ProDave

It's generally only ones that have a pumped effluent output that have an alarm to let you know if the pump has failed and the holding tank is over full.

#3 jsharris

Ours came with an alarm. The main sensor is a pressure sensor on the air pump, to warn if the aeration stops. Seems a good idea to me, as with no alarm you'd have no easy way of knowing if the pump had failed. The alarm box has a battery back up and both sounds a (rather quiet) beeper and flashes a strobe light fitted to the top of the cover. The only slight problem is that the pump failure (low pressure) alarm goes off whenever there is a power cut. The beeper isn't audible enough to be a nuisance (or to be useful, TBH), but it would have been a good idea to not have the alarm go off just because there's a power cut, in my view.

Edited by jsharris, 23 March 2016 - 07:00 PM.

#4 PeterStarck

I have a WPL plant which has a visual alarm (LED lamp) on top of the air pump box which indicates if the air pump is not working, usually if a diaphragm has broken. I have also installed another LED lamp to warn if the liquid pump has failed that pumps effluent up to the reedbed. Although they are IP66 they don't seem very reliable. I am looking into installing a wireless audible alarm which would sound in the house. I want two channels one for each of the lamps sending to a single sounder in the house.The only suitable system I have found so far is around £100, so I am still looking. I have a friend who worked in electronics so may ask him if I can't sort it out.

#5 jsharris

I've looked at doing this, too, but decided that I'd just relocate the alarm unit to a position where it's more visible.

I wanted to build a small stone box at the top of the drive to contain salt, as the bottom of our drive has a steep slope and even a thin layer of snow makes it impassable. Whilst sketching this up I realised it was just as easy to build a stone box with two separate compartments, one to hold a supply of rock salt, the other to house the pump and alarm unit. I've dug a trench down from where this will go to the treatment plant and put a duct in, the intention being to run a power cable and the pump air line up from the treatment plant to this small box. As it will then be right by the side of the drive, just behind where I park my car, I'll be able to see the alarm light far more easily. I can also see it from the front windows of the house.

I think there's also an added advantage in that it will be a lot easier to work on, or replace, the pump when it's relocated, as there will be no need to go down to the treatment plant and unbolt the lid to get at the pump. I'm about to buy a spare pump and a couple of spare diaphragm kits, as I understand that it's usually just the diaphragm that fails on these things. I have also looked at the smaller pumps that have no diaphragm, just a free-floating oscillating ceramic coated piston. These are reputed to have much greater reliability than the diaphragm pumps, but do seem to be a fair bit more expensive for the same pumping capacity.

As this new stone box will have a mono pitch roof that faces south, I've also looked at making the treatment plant solar powered. I can just fit a 200 to 250 W panel as the roof, and that would probably be enough for the 30W needed for the pump and maybe another watt or two for the alarm. Ideally I'd need a DC powered pump, though, and so far I've had no luck in finding one.

#6 Triassic

I used to work on a scout camp site and one particular septic tank had a pumped discharge, because of its remote location an alarm was not feasible, so we fitted a float in a tube, as the level rose in the tank the float pushed a stick with a sign on the end up out of a box, the sign said "tell the warden the pump has stopped". It worked!

#7 jsharris

I think the main problem isn't the level in the tank, but whether the aeration pump is working. These usually need to have their diaphragms replaced every two or three years, and without an air pressure failure alarm you'd have no way of knowing whether or not the pump was still doing it's job.

Our unit has a lid that's bolted down and secures both the top lid that keeps the weather out and the lift out chamber that contains the air pump, alarm and power connection. It's not a very good arrangement in my view, as it's not that easy to bend down and work on this stuff in what amounts to a box set into the top of the tank itself. There's a bit in the instructions that says that the air pump can be mounted remotely, and some other tanks seem to do this with no problem, so my idea is just to feed power and a bigger air pipe (to ensure low losses) up to the remotely located "pump house". The alarm senses pressure at the pump outlet, so moving that as well will be easy.

I spent an hour or two looking at ways to make this solar powered, really just to get rid of the 30 to 40W constant load, 24/7. 365 days a year. That load amounts to around 350 kWh per year, not much, but around 10% of our total electricity usage, so worth saving. A 260W solar panel would be a good fit as a roof over the 1.5m wide by 0.8m deep stone box and is in an unshaded area, facing south. PVGIS tells me I should get around half the energy we need over the year from this panel, so I'd need to use some grid power as a battery top up. It might well be cheaper and easier to forget the extra PV panel, and just use battery pack running an inverter to run the pump, with the battery charger using a combination of excess PV sensing from the house array and battery charge state to initiate a mains powered charge. As we generate around 6,000 kWh or so per year and always have a fair bit spare to export, this should work to reduce the mains demand from the pump. Whether the cost of the battery pack, charger and inverter is worth it is debatable, though.

I've read that some systems can run quite happily with the aeration pump turned off overnight. I need to do some more digging around, as if this is the case with ours then I could cut the standing energy use a lot by just turning it off at night with a time switch. The saving would be over 50%, as for much of the year the pump would be running from excess PV generation anyway during the day.

#8 PeterStarck

Our air pump is housed in a separate box which is remote from the sewage treatment tank. It is partially set into the ground and has the warning light on top of the box. We were told our tank didn't need emptying for around four years. This turned out not to be true for us and after two years or so the back pressure on the air pump caused one of the diaphragms to split. I now fully service the pump every two years and have the tank emptied every year. We've not had a problem since, better safe than sorry!

#9 jsharris

Thanks for the tip, Peter, as we were also told the same thing! Yesterday I ordered a spare pump and two diaphragm kits, so I can quickly swap the pump over then replace the diaphragms in the other pump and keep that as a spare.

It's interesting that the cause in your case was additional back pressure, presumably caused by the sludge layer getting to be above some of the aeration holes. It makes me wonder whether a better alarm, using a pressure sensor rather than a pressure switch, might be a good idea. It would be very easy to do, and having something that indicated increasing pump pressure could be a useful early warning system. Knowing how long it takes for the sludge level to rise and cause the pressure to increase would be useful, as it would give a better idea of when the tank needs pumping out, which could mean increasing the pump-out periods to better suit the usage pattern.

I shall put my thinking hat on, as I don't like the alarm system we've got, anyway, so this would seem to be a good excuse to design one that has a bit better functionality. It should be pretty easy to integrate a low power radio link into it, too, so that data could be sent to an indoor indicator unit. It's certainly easier to do this as a part of the overall alarm system design than it would be to make some sort of add-on unit.

Edited by jsharris, 25 March 2016 - 05:02 PM.

#10 PeterStarck

I think that because the aeration compartment was completely full of solids the pump wasn't able to provide air and the weakest point was a diaphragm. My pump is a Secoh EL-100S which provides 100l/min and is rated at 115W. Secoh have now replaced that pump with the JDK100 which also provides 100l/min but is rated at 65W, a good saving, but at around £200 is not cheap.

#11 jsharris

Ours uses the Secoh JDK60 pump, which we opted for at the time as an energy saving over the standard pump, which would then have been the Secoh EL 60. I seem to remember there was a small premium for choosing the lower energy pump, but not much.

I've just bought a new JDK 60, plus two sets of diaphragm kits as spares, for £196, including VAT and delivery. I had a look at the pump curves and there is a distinct difference between the older EL series and the newer JDK series, it seems. The EL 60 that our BioPure 1 would have come with (had we not opted for the BioPure 1E with the lower power pump) has an air delivery that ranges from 105 litres/min with no back pressure, to zero at about 0.34 bar, as a straight line. The JDK 60, which is supposedly a direct replacement, delivers 130 litres/min with no back pressure, decreasing to 60 litres/min at 0.2 bar, in a straight line, then decreasing further to zero flow at 0.3 bar.

I'd like to understand how much aeration is really needed for the system to work, as I have a suspicion that treatment plants probably over-aerate in order to err on the side of caution. As the power drawn by the pump is a considerable amount of energy over the whole year, and as some suppliers say their units can be safely turned off for a few hours a day, I can't help but wonder if there isn't a way to control the air pump more efficiently. I haven't looked into it in detail, but we do have a ORP meter for testing our well water (it measures the oxygen level in the water, indirectly, as the Oxygen Reducing Potential. If I knew what ORP was required in the treatment plant, in order for aerobic bacteria to thrive, then it should be possible to monitor ORP and use that to control a timer that is sequenced to keep the level optimised for best operation. My guess is that the pump would not run for anything like 24 hours at a time for most people, who may only give the unit a light load (there are only two of us, for example, yet our unit could deal with 5 people, so is probably oversized already).

#12 PeterStarck

The tank air inlet on our system is at the bottom of the aeration compartment and that compartment is around 1.6m deep. From experience I know that when it is full the air pump is at it's limit. Now I am having it emptied more frequently perhaps I could use a smaller pump. I thought about running the pump on and off on a one hour cycle but was concerned that when the tank is quite full, on startup, the diaphragms would be more stressed.

#13 jsharris

That's a good point about start up stress that I'd not thought of. If sludge settles over the aeration holes then it probably would push the initial pressure up. In theory the pressure at 1.6m down should only be about 0.157 bar, well under the stall pressure of around 0.3 bar, so it all depends on how viscous the sludge is after, say, 8 hours of settling. In our case I don't think that we'd have the same problem, as the Biopure seems to use an air lift system to lift sludge from the bottom, using a draft tube with the air injection point some distance from the very bottom. In essence they use what looks to be a classic air lift, as I used to pull sand from the bottom of our well, and this will work well with the air injection point well above the base of the suction tube. I think it's probably unlikely that the sludge level would rise as far as the air injection holes unless the unit needs emptying, so monitoring pressure for us would be a good way of telling when the system needs pumping out, as I suspect we'd see a step increase in pressure over the space of a few days. I've found a cutaway digram that shows how this air lift system works:



Out of interest, what is the capacity of your system? I'm curious because our 1 to 5 person unit has a much smaller pump than yours. around 35 to 40 W (ours seems closer to 30W when I measured it, but I suspect that's because, like yours, it's only working at around 0.16 bar, rather than the 0.2 bar that they seem to use as the standard to measure these things by).

It would also seem possible to run these pumps from a variable frequency inverter, to reduce the pumped volume and hence power. At the moment they work at 25 Hz (half mains frequency) by the simple expedient of running the coils via a diode to halve the effective frequency (nasty as far as the supply goes, though, just pulling power from half the waveform). It should be possible to run them at any frequency from around 5Hz up to 25Hz I'd have thought,

Edited by jsharris, 26 March 2016 - 11:25 AM.

#14 PeterStarck

According to the user manual the sludge capacity is 1500l. It states "Check the percentage of settled sludge (i.e. sludge volume) to the clear liquor. The optimum level of solid settlement is normally between 5-50%. Whenever the sludge volume reaches 70%, the plant should be de-sludged." Not so easy estimating those volumes! From your diagram it looks like your system is very similar to mine. Your diagram shows a very small amount of settled sludge but in practice after running for a year or so I would have thought the level would be above the bottom of the air inlet if allowed to settle i.e. turned off. I think only by continuous running are the solids kept in suspension.

#15 Alphonsox

I have just checked the pump that our 6 person Kingspan unit is going to use - it's a Secoh EL80. This wasn't what was specified but its been over a year since purchase. Looks like I will need to replace this with the JDK version before installation.

#16 jsharris

Peter, your unit looks very similar indeed to ours, even down to the dimensions, ours is 1.8m in diameter at the top and about 2.25m from top to bottom. I'm sure you're right about the need to keep solids in suspension, but it would be interesting to know just how much air is needed to do that, together with the amount of air needed to keep the bacteria happy.

The obvious question then is, why does your unit need the much more powerful EL-100/JDK-100 pump, when ours seems to work fine with the smaller JDK-60? There seems to be a significant price difference between the two pumps, and clearly a fairly big running cost difference. It does make me wonder how much real work has gone into some of these systems in terms of sizing the components.

I hunted around on line for the price of a spare pump and spares, and the best combined deal I could come up with was from Direct Drainage. The SE24 diaphragm kit (which also fits your larger JDK-100) was £17.00 and the JDK-60 pump was £120.00, both prices plus VAT and delivery for the whole package was £8.00. Hopefully I have enough spares to keep the system going for several years now, but I am very interested in monitoring it to see if I can spot indications of impending trouble. I've looked at making a better alarm unit, but the cost of a decent pressure sensor seems to be OTT for what is needed. Instead it seems a great deal cheaper to just fit a simple pressure gauge (around £2 to £3 from ebay) tapped into the air line. I'll have the advantage of having our pump right behind where I park my car, at the side of the drive, so keeping an eye on the gauge for signs of the pressure increasing will be straightforward. If your unit is some way from a convenient monitoring point you could look to see if running a small air pipe tee'd off from the pump is practical, so you could locate a gauge where it could be easily seen.

#17 PeterStarck

When I have finished the building work I will spend sometime tinkering with various areas of the build that could be improved. This is certainly one area where I think significant energy could be saved by fitting a more efficient and or smaller pump and incorporating an improved warning system. My tank has been in use since 2010 and I will check whether the manufacturers have improved the design in any way since then. Thanks for the input, if you complete any improvements I would be interested to know the effect.

#18 PeterW

Could you not use a pump that has some sort of accumulator so that when it switches on you get a burst of higher pressure air to clear any nozzles or air inlets ..??

Not 100% au fait with these systems so tell me if I'm talking rubbish ..!

#19 jsharris

Possibly, but the pressure is pretty low. The maximum pressure that the pump will deliver is only about 0.3 bar or so, around 4 to 5 psi. When run at the maximum pressure, as Peter S has found out, it seems that the diaphragm in the pump may fail. My guess is that it may well need a fair bit more than the 0.3 bar that these pumps are capable of delivering to clear air holes that are blocked with sludge.

There are pumps available that don't have a diaphragm, but use a linear piston with a ceramic coating. I'm using a small one of these to pump ozone into our water supply aeration system, as they have a reputation for having a very long working life, but I've not seen this type of pump used on a sewage treatment plant and they still suffer from the same problem of not having a high maximum pressure. IIRC, the Hailea piston pump I have will only pump to around 0.2 bar, so lower than the diaphragm pumps. I have a feeling that most of the pumps used in sewage treatment plants were originally designed as aeration pumps for fish ponds and have just been adopted by the companies making the treatment plants.

#20 ProDave

Just to add to this, we are using the Conder HDPE ASP6 http://www.waste-wat...ASP-6-20PE.html

Another incarnation of pretty much the same design of air blower treatment plant.

According to the installation instructions, ours has a "130W max" pump that can either be mounted inside the unit or remotely.

I must confess I haven't looked at the rating plate on the pump yet, I merely opened the lid to make sure one had been supplied, and put the lid back on.

 

 

Following on from the above, I decided it would be nice to have our alarm relocated with an indoor display, that would not only provide the "low pressure" and "high level" alarms at a convenient location, but with the addition of an indoor pressure display, to give early warning of sediment build up, so avoiding (hopefully) the problem that Peter reported.

I have the pressure sensors, have tested them and the interface to a wireless data link that will comfortably work for 50 to 60m through walls.  I have three choices.  I can make the outdoor unit larger and fit a battery a solar panel, and have the indoor unit as a hand-held battery powered unit that could interrogate the external unit on request.  The power demands make this arrangement impractical for giving a real-time alarm, it would rely on being interrogated to find out whether things were still OK, its major shortcoming.

The next option is to have a solar powered external unit transmitting to a mains powered internal unit.  This would provide continuous monitoring and alarm, at the cost of a solar powered external unit that would be relatively large (because of the battery and solar panel size) an would cost more, plus the battery would need replacement every few years.

The final option is to have a mains powered unit inside and outside the house.  This would probably be the best solution if there is mains power available at the treatment plant, as it would be the most reliable.  In terms of power, it would use less than 1W at each end.

This post is really to see if anyone is interested, as I'm making one for our build, and it's very little trouble to make one or two extra.  I'm probably going to opt for mains powered at both ends, as that suits us better, but there really isn't much to choose in terms of hassle to make other types.

Anyone interested?

[Calvinmiddle]

This is something I would be interested in, we have the bio pure as well, but we have remotely placed the pump from the unit as we only have a small area of grass so have got a manhole filled with soil and turf that can be lifted to get to the main treatment plant.  Can't remember the power of make of the pump, will have to check it but the pressure sensor and alarm was a small battery power unit that came with rechargeable batteries and a mains adaptor.  These are sited a couple of meters away from the plant and the air is pumped via a clear plastic pipe reinforced with a spiral wired running through the middle of it.  The pump is protected from the weather under a box made from feather edge boards to resemble a small bee hive.

Will take a couple of photos to explain better.

[Jeremy Harris]

Sounds very much like the arrangement I'm putting in place.  Currently the pump and alarm are inside the unit. under the lid in the internal equipment box.  This works OK, but is pretty difficult to get at, so I'm in the process of building a concrete base that will have a small stone box built on it, with two compartments, one for the air pump and alarm unit and the other as a silo for some rock salt, as our drive is so steep that the slightest bit of snow makes it impassable.

I don't like the idea of relying wholly on the external alarm (and mine seems much the same as yours, a plastic box with a rechargeable battery and a beeper plus a flashing light), as there is a good chance that we may not see it for some time.  Also, as Peter noted above, the back pressure on the pump seems to build as the sediment layer gets deeper, so something that indicates the air pressure seems useful, as you can watch for an increase and use that as an indicator as to when to order a pump out.

I've bought a bag of inexpensive pressure sensors, so it's pretty straightforward to just transmit the pressure to the internal box and then let the internal box handle the alarm sensing.  In our case, we have the pumped outlet version so there is a second alarm feature that senses when a level switch has detected an over-level event, signalling an alarm that the emptying pump has stopped working.  It'd be easy to monitor a level switch as well and transmit the state of that as a part of the data transmission.

[Gone West]

Hi Jeremy, yes I'm interested in the setup using mains at both ends. Our box containing the pump etc is semi-underground. If the transmitter is in the box does it need an external aerial? The box is around 8m from the house.

[Jeremy Harris]

I can easily fit an external antenna, but 8m is pretty short range so the antenna inside the plastic box I'm looking at using should be OK.  I've tested the basic data link and it looks good for well over 50m.  With a bit of luck this weekend I'll get the first pair of units built so I can start testing.  I can emulate your case by just fitting the alarm transmitted down in the equipment box of our present set up, as that's inset down into the top of the tank so partially below ground.  If it works OK from there to the back corner of our house, in the utility room where I'm going to fit the alarm and display, about 25m away, then it should be fine for your case I think.

The parts are pretty cheap, its things like the weather tight plastic housings that are the most expensive parts.  The internal display and alarm will, I hope, just fit into a standard socket wall box, I'll try and fit it in a single, but it may need to be a double.  What I've done in the past is to machine a hole for a display into a blanking plate for a standard box.  This seems to give a neat appearance and matches well enough with other fittings.   It will have to have a lead to a 13A plug with the appropriate fuse, though, just to be "portable" and avoid the need for wiring regs stuff.

[Guest Alphonsox]

Definitely interested - as discussed above we currently have no failure indication at all.

[Jeremy Harris]

Currently I have enough parts to build 4 units, so on the basis of first come, first served, I'll start with a batch of four for Calvinmiddle, Peter Starck, Alphonsox, and the prototype for us.  Cost is probably going to end up around £25 to £30 at a guess; I haven't added up the cost of the parts yet, but will do over the weekend.  I think it's unlikely to be much over £30, worst case.  It could probably pay for itself fairly quickly if it prevents an early pump failure from monitoring the pressure, or allows optimisation of emptying periods (i.e. don't empty the unit until the pressure starts to creep up, perhaps).  I'll stick the design on here as an open project, so that anyone with the interest and a bit of experience could build one.

The main reason that such a thing is affordable now is the advent of mass-produced home blood pressure monitors, as pressure sensors have always been far too expensive in the past.  I'm using a small and very cheap pressure sensor that's intended to be used in a blood pressure monitor and only costs less than £2.  The test sensor I was using for development is one scrounged from a bin at work back in the late 80's that, IIRC, cost around £400 or so.

The display will be a large character 8 digit LCD, with gentle backlight, that will normally just display the air pressure, in bar, at the pump (so this alarm units sense pipe will need to be tee'd into the air pump outlet, as those will existing alarms will know).  I can fit a beeper to the internal unit or just rely on a bright red LED flashing away, with the fault condition showing on the display.  I'm in two minds about adding the beeper at the moment.  It's cheap (around £1) and easy to do, but it could be a nuisance if it starts beeping in the middle of the night.

The treatment plant end will have an input for a level switch, simply because I already have a level switch on our system, but this can be ignored for those who don't have pumped systems.

[Gone West]

I would be more interested in a beeper than a flashing light. The beepers I've looked at before were over 110dB which is too loud.

[PeterW]

Jeremy - did you intend to use a PIC to do this ..?

Just wondering if it forms the basis of a "monitoring" ecosystem that could take a range of sensors..?

[Jeremy Harris]

I agree, Peter S, a "gentle" beeper would be fine, just enough to get your attention without being ear-splittingly loud.

Peter W, it is PIC based, as are all the sensor systems in our new build.  I have data being logged to an SD card every 6 minutes from a load of temperature sensors around he place.  I also have a portable stand-alone logger that monitors temperature, relative humidity and CO₂ levels, and logs that to a µSD card every 6 minutes (some here have borrowed it).  The excess PV diverter that's been running for a fair time now, initially switching power to an immersion, now doing the same to the Sunamp PV is similarly home made and runs using an 868MHz data link from the meter box outside to the internal switch unit.  The same unit "broadcasts" import/export power levels every 10 seconds to any receiver in the house, so that it's easy to see the current state.  Yet another of these devices runs my electric car charge points (an EVSE in EV parlance) and generates the required signals, and interprets them, for J1772 protocol control.  I currently have four home-made EVSE units, ranging from micro-miniature ones built inside J1772 connectors, to the proper boxes at the new house, one with a built in energy meter for my car normal charging and another "guest" 32A charger on the garage, that accepts the standard IEC62196-2 cables that most EVs are supplied with for use with public charge points.

At any one time that are at least half a dozen or more microcontrollers working away in the new house!  The basic logger is in this blog post:

 which also has a photo of a remote energy monitor that picks up the 433MHz "broadcast" from the main outdoor energy metering system.

[Gone West]

The input for a float switch would be very useful for me too. I have a separate containment tank and I fitted a float switch and warning light to indicate if the pump has failed, so I could connect that to the transmitter as well. I appreciate your effort in this and would like to thank you for your help.

[Jeremy Harris]

Thanks Peter, the float level switch input will be there on all units, so if a switch is connected (a normally open one that closes when the level is too high) then that part of the alarm works as well, if the switch isn't fitted the unit just ignores it.  It's about 10p to add a connector for a float switch to the unit, the rest is just code, which is free.

[Calvinmiddle]

So here is the set up I have on our treatment plant.

This is the corner it's in, you can see the manhole in the grass as the soil has dried out and the metal edges are showing a bit more than normal.  The pump is under the beehive.

This is the cover off and you can see the pipe running through a drainpipe to the unit.

Finally this is the beehive removed and you can see the pump (50w), the alarm and the power sockets.

[Jeremy Harris]

That's similar to the arrangement I'm building at the moment, with a double waterproof socket, and our pump is the same Secoh model.  Our alarm is currently a much bigger unit, with a level switch input and a strobe light output, with a sealed lead acid back up battery.  The alarm sender unit I'm currently building is smaller than the unit you have, and normally just has a mains power lead plus the small pressure sense pipe, but can have a level switch added very easily.

I've just finished testing the outdoor unit hardware and firmware and it seems to be working fine, transmitting pressure and the status of the level switch every ten seconds to the indoor unit.  I spent around 2 hours trying to get the pressure sensor working, only to realise that the (Chinese..........) data sheet was mis-labelled, and the connections were a mirror image of those shown.  It's working fine now though, and reads from zero to 0.40 bar, which should be more that enough, because the pumps only normally deliver around 0.20 to 0.25 bar when the pipe's blocked.  The normal operating pressure should be around 0.12 to 0.15 bar for most tanks, I think.

[Jeremy Harris]

Right, I made a start at the weekend (as light relief from collating VAT receipts) and made up a prototype.  It's very much a prototype, and although I may end with it installed on our system, there are probably things I will change for "production" versions, just to make them a bit easier to make, and a bit neater.

First off, this is the small waterproof box that goes at the treatment plant end:

I should have added a scale, but the air pipe to the pressure sensor is about 5mm in diameter and the case is 65mm wide x 115mm high, so pretty small.  This needs to be mounted vertically, with the transceiver antenna at the top.  Ideally it needs a reasonably good line of sight to work, but I tested it yesterday with this unit sat inside the garage on the floor (our garage is right at the opposite end of the site to the corner of the house where the display unit will be) and it worked perfectly (that was a range of around 35m through several thick walls).

This unit has a small pressure sensor that measures the air pressure and transmits it to the indoor unit.  It also has a pair of screw terminals to allow a "normally open" float switch to be added if needed.  The unit works with or without the float switch, as it all it does is send data on request to the indoor display and alarm unit.  I'm probably only going to make some minor revisions to the design of this unit, just to make fitting the cables and pipe a bit easier.  The green LED is just for local diagnostics.  You can only see it with the cover off and it has a sequence of flashes that indicate normal operation, a fault condition and what that might be and also a confirmation of sensor zero calibration.  There should be no need for zero calibration in use, as the sensor is very stable, but there are variations from one sensor to another in the zero offset, so I had to build in a way to calibrate this out.  The red link to the left of the LED is the calibration link.  If this is removed (with the unit powered off and the air pipe disconnected) and then the unit is powered on, it will measure the zero offset of the sensor, store it in non-volatile memory (should last decades) and then give three quick flashes of the LED to show that calibration is complete.  The power can be turned off, the red link replaced and the unit then installed and operated.  This isn't something I envisage anyone needing to do, though, as I'll do the zero calibration as a part of testing the functionality.

The indoor unit has a small display to show either the air pressure, or to indicate "HI LEVEL" if a float switch is fitted to the treatment plant.  In normal use the unit just checks and displays the pump pressure about every 5 seconds, and looks like this (I may change the case, as I'm not 100% happy with this fitted in a patress box and working from a plug-in wall supply):

The red LED on the front comes on continuously when there is an alarm condition and an internal beeper sounds a short beep every 5 seconds to alert you to the alarm.  If the alarm was triggered by the float switch operating, then the display will show "HI LEVEL" instead of the pressure, the LED will be on and the beeper will beep.  If the alarm was triggered by the pressure dropping from the pump, then this will show as a low reading on the display.  I currently have the pressure alarm threshold set to 0.03bar. so the alarm operates at or below this pressure and the unit looks like this when the pressure reaches the low pressure alarm threshold:

One advantage of being able to see the pressure all the time is that you should be able to get a feel for the normal operating pressure for your unit (the pressure above was not from our unit, I simulated it with a bit of silicone pipe and a syringe).  If the pressure rises, then based on Peter's experience of this being an indication of sludge build up, that can damage the pump diaphragm(s), then it may well be a good indicator that the unit needs emptying.  If the level in the tank stays pretty constant, then the pressure will, too, as the air pressure depends very much on the head in the tank.  This means that for some systems there may be a slight pressure variation as the tank fills to a slightly higher level before draining to the normal operating level.

The good news is that everything works as I'd hoped and that the radio link is very robust.  It's a two way link, where the outdoor unit sends a very short "wake up" data burst saying that it's ready to transmit data, the indoor unit responds by transmitting a "ready to receive data" message, the outdoor unit sends the data and the indoor unit transmits back confirmation that the data has been received.  This makes the data link pretty robust, and combined with the low data rate used and the high receiver sensitivity means that the range is likely to be well over 50m in practice, and it will easily cope with being obstructed by walls.  It's helped by having a decent antenna on the indoor unit, and for a difficult location it would be easy to add a similar antenna to the outdoor unit, rather than rely on the small helical one inside the box you can see in the photo.  The downsides of fitting an external antenna to the outdoor unit is that it would need some additional sealing around the antenna connection and it would almost double the vertical size of the unit.  The outdoor unit is in an IP65 rated box, so could be fitted to a small post next to the treatment plant to get it above ground level, if that was found to be needed.  My view, based on the tests I did yesterday) is that it would probably work OK, with a range reduced to maybe 25m or so, even if it was fitted in the pump compartment of a typical treatment plant.

I have a few pressing things to do over the next week or two, but I'm then planning on looking at revising the mechanical design a  bit, but leaving the firmware pretty much as-is, as it seems to work fine.  One thing I'd like to do is get rid of the plug-in 5V power supply for the indoor unit, and make it either a wired in unit (means adding a fuse internally) or a plug in unit with a 3 A fused 13A plug.   One no-cost option is to fit a display that allows more data to be shown, perhaps in a slightly different case that doesn't need me to use the milling machine to make the display cut out, so the pressure could still be displayed with the level alarm, or a high pressure warning (rather than an alarm) could be displayed if needed.  It's dead easy to do these sort of changes, literally ten minutes work, if that to change the code and test it, so I'm welcome to any ideas for improvements.

[Guest Alphonsox]

Looking very good. In my case I may have to remote mount the indoor aerial, I have real problems getting radio signals through the foil lined PIR insulation in the place.I'm not sure what effect this will have on 433MHz signals but it stops mobile signals almost entirely.

[Jeremy Harris]

It may well work OK, as our house similarly blocks mobile phone signals completely, but these short range radios seem to have no real problem going though the wall or windows.  The advantage they have is that the lower frequency is better at getting through buildings, they use a very low data rate that extends the range for a given power to maybe three or four times that of a mobile phone and they have a better antenna and almost certainly a more sensitive receiver (mobile phone receivers have to deal with poor built-in antennas and lots of digital noise from the phone circuitry).

If it is a problem for your location, then adding an external antenna to the outdoor box will just about double the performance, and is probably the easiest thing to do.

[Calvinmiddle]

Looks great Jeremy

[Jeremy Harris]

I've found some small wall-mount boxes that come with a display cut-out for a good price, so have ordered a few to see what they are like (they aren't expensive).  I'm hoping that I can fit a small power supply inside (the same as the small 1 W, 5 V supply in the outside unit, the orange block at the lower left in the photo).  If so, then it increases the price by a couple of pounds, but makes for a neater arrangement, as I can fit a fuse to the unit and then it could be either hard wired to the house supply or run from a 13 A plug.

One thing about the case I've found is that it takes a bigger (but cheaper) display, so I can display more information, rather than just the pressure.  It means I could display the high level text warning and still display the air pump pressure, for example.  It also means I can backlight the display, as the existing display wasn't available with a backlight, and I think it would be useful to have a not-too-bright backlight to make the display more readable.

If anyone has any preferences one way or the other, then let me know, as I have parts (except the cases that are on order) to make either version.

The air pressure sense pipe on the outdoor unit will be a 4mm diameter barbed fitting, threaded into the lower part of the case, next to the power cable and the cable inlet for the float switch (if needed).  Hopefully this will be OK for whatever pressure sense pipe you may already have, but if not I can get hold of tee fittings from the aquatics place over the road, they seem to do one with a 19mm (3/4") straight through and a 4mm side fitting, that looks very like the fitting that is already on our air pump.

I've done a quick estimate of the parts cost and it's a little more than I'd hoped, at around £36.  I had hoped to be able to make them for under £30, but the cost of all the little bits seems to add up.  Please say if this is a bit too pricey!

[Guest Alphonsox]

A bigger, backlit version would be my preference - The price is still far too cheap !

[Jeremy Harris]

The bigger backlit version is my preference, too, as I think it will be easier to read where I'm going to put it (on the utility room wall, where the thermostat receivers, ASHP command unit etc are located) and I like the idea of hard-wiring it to the supply, just for neatness.

Price is an odd thing for things like this, as when you're knocking stuff together you sort of get used to things being much cheaper than an equivalent commercial device, even if you can buy a commercial device that will do the job.  I think I added up the cost of the excess PV power diverter I made and it came to around £25, that's for a unit where the power sensing is done outside, at the incomer, with a radio link to both the power switch for the Sunamp PV and to a plug-in power display that receives the data and has a display showing whether the house in importing or exporting, and what the instantaneous power is.

I tend to think that spending more than £30 on any little gadget like this is getting to be too much!

[Calvinmiddle]

I don't really need the float level, but for the sake of a couple of ££ I'll go with with the same as everyone else to future proof it and keep it easy for you so there is only one version.  Regarding the power source I'd prefer a plug in

[Stones]

This makes very interesting reading.  I think I'll be looking for one of these alarms in due course, as I'm planning either a Bio-Pure or Conder aeration plant. Appreciate I missed the boat on being part of the initial project, but if it can be easily replicated by someone with basic electronic and soldering experience / skills, then that would be ideal.

[Jeremy Harris]

54 minutes ago, Calvinmiddle said:

I don't really need the float level, but for the sake of a couple of ££ I'll go with with the same as everyone else to future proof it and keep it easy for you so there is only one version.  Regarding the power source I'd prefer a plug in

They are all identical, float alarm or no float alarm even down to the code being the same, the only cost difference would be that I could leave out the terminal block for the alarm connections and save a bit under 10p.

If you don't need the float then just don't fit a float switch and the unit will just assume there's no float alarm.  If you need any other sort of alarm, then you can wire any normally open switch to the float terminals if you want, and when the switch closes it will sound the alarm inside the house.  If you want to do this, then let me know and I can change the code to display a message other than "HI LEVEL" on the alarm unit display.  You could wire this to pretty much anything, like an anti-tamper switch, a switch on the door of a chicken shed to remind you that the hens are abed (did one of these for a friend years ago, so she could check to see whether she'd remembered to shut them in at night without needing to go outside with a torch to double check).  Anything you can run a bit of two core cable to with a normally open switch could be used to trigger it.

[Jeremy Harris]

31 minutes ago, Stones said:

This makes very interesting reading.  I think I'll be looking for one of these alarms in due course, as I'm planning either a Bio-Pure or Conder aeration plant. Appreciate I missed the boat on being part of the initial project, but if it can be easily replicated by someone with basic electronic and soldering experience / skills, then that would be ideal.

I dare say I can add an extra one Jason, not really an issue as I'll now have bits for 5 units, rather than 4.