Our house is 12 months old and the MVHR is working well. However, although the unit is fairly programmable (rates, temps etc), it's not something you'd want to do daily, or even monthly. (Well maybe monthly).

I have loads of temperature monitoring through the house and can see the 4 temps in the ducts to/from the house. So I can see when the bypass switches in, and can see the effect.

Let's say I think the effect could be better. I think loads of air is bypassing the bypass, but we'll overlook that and continue ... what I'd like ideally is more manual control of the bypass. In hot summer weather I'd like the bypass open all night to get maximum cooling, instead of closing again when inlet temp drops to 21C.

Yes I know I could change the bypass temp, but perhaps there is a way I could more easily control the bypass either via a controller, or a connection to the unit?

I'd be grateful for a tip from anyone with experience - thanks ... David

I have a Vent-Axia Sentinel Kinetic BH (summer bypass + humidistat) and have been vaguely pondering better control options. At present, my only external control input is a volt-free connection from the cooker hood that boosts performance when the cooker hood fan is running.

There are two free 0-10V connections (P1, P2) which can be configured for humidity, CO2, or temperature external sensors, However, it isn't entirely clear from <a href="http://www.vent-axia.com/files/pdf-downloads/439817%20S.pdf">the documentation</a> how they interact with the control program. In particular, whether the external temperature sensor overrides the internal one on control of the summer bypass.

There is also a full Electronic BMS (Building Management System): <A HREF="http://www.vent-axia.com/product/clone-sentinel-kinetic-bh.html">A full range of two-way digital signals are available through the RJ11 connector on the control board. The BMS system provider will translate this signal to extract the desired data. Contact Vent-Axia to discuss your specific requirements. </A>

When time and funds permit I would like to try to use an internal air quality monitor like the <A HREF="http://www.nano-sense.com/">Nanosense</A> <A HREF="http://www.nano-sense.com/E4000.php?lang=en">E4000</A> and <A HREF="http://www.nano-sense.com/P4000.php">P4000</A> to provide control information.

I have an Ethernet connection next to the MVHR unit so it might be feasible to knock up something relatively inexpensive with a <A HREF="http://www.raspberrypi.org/">Raspberry Pi</A>, <A HREF="http://www.arduino.cc/">Arduino</A> or even <A HREF="http://www.netmf.com/gadgeteer/">Microsoft .NET Gadgeteer</A>. I haven't yet contacted Vent-Axia to try to find out more about that interface. Note that I am not an electronic engineer so this may be non-trivial!

My other option (as I plan to use EnOcean elsewhere) would be to use a <A HREF="http://www.thermokon.de/EN/easysens-31/srcao-multi-multi-actuator.html">Thermokon SRC-AO Multi</A> to drive one or both of the 0-10V inputs. Fortunately, Vent-Axia provide the necessary 24V input.

I will probably investigate improved heating controls before the MVHR so it may be some time before I get around to it. I would be very interested to hear of any other people's experience with controlling MVHR units.

I contacted Vent-Axia and was sent a copy of a document "Sentinel Kinetic Range BMS Capabilities". An RJ11 to RS232 cable is available which supports 9600 baud, 8 data bits, 1 stop bit, no parity. Current software just supports stop and start of the unit via this serial interface. Access to internal data would require software changes. It would have been handy if the internal sensors (temperatures, RH, motor speeds, switch states, ...) could have been queried and control parameters adjusted.

Looks like the two 0-10V inputs offer the most flexible of the currently available options.

Posted By: john_connettI contacted Vent-Axia and was sent a copy of a document "Sentinel Kinetic Range BMS Capabilities". An RJ11 to RS232 cable is available which supports 9600 baud, 8 data bits, 1 stop bit, no parity. Current software just supports stop and start of the unit via this serial interface. Access to internal data would require software changes.

I don't believe that. The temptation for programmers to export information like that is very strong. I think it's more likely that they simply haven't documented such things in order to avoid the headache of supporting them. There are any number of devices with undocumented RS232 diagnostic interfaces in existence.

Posted By: PeterWI'm always amazed at manufacturers of system components that invariably don't talk to each other or use 'proprietory' protocols that excludes innovation.

There are a lot of social reasons why this sort of thing happens. E.g., as Andrew_Doran says, there's the support issue. Answering half a dozen calls/emails about the interface could easily wipe out the real commercial benefit of supporting it.

Also, people working on these things don't think of them as “system components” but rather as the centre of their own little universes. Happens all the time with all sorts of components: charge controllers, inverters, CH systems, fridges and freezers and so on.

One of a number of reasons why I plan to make my own heat-recovery system, though the main one is the ridiculous costs; 10 × sane component costs it seems to me.

Posted By: PeterW: “I'd be interested to see what you are planning Ed”

Wrote a long post explaining the whole thing in a new thread then the forum software said I wasn't signed in and threw it all away so for now I'll just say I'm thinking of something like this:

http://www.builditsolar.com/Experimental/DryerHX/DryerHXTest1.htm#HX%20Description

but about 3 metres tall running down the outside of a gable wall but inside an attached porch/greenhouse thingy.

Slightly old overall sketch of the house: http://edavies.me.uk/2012/01/house-sketch/ As always, the design continues to evolve (and will, I hope, continue to do so even post construction) but that's fairly accurate for these purposes.

What I have in mind for the MVHR is a sandwich of twin-wall polycarbonate running vertically down the gable wall in the porch/greenhouse, just to the right of the front door with the fans up at the top blowing/sucking through holes in the gable end to ducting in the loft area inside the house.

Inlets and outlets at the bottom initially connected to the outside world (via ducts under the porch deck or along the bottom of the wall/deck join to the outside on the south side) but something I want to experiment with is exchanging air for inlet, outlet or both with the greenhouse depending on temperature and humidity conditions.

Incoming air runs upwards through the channels in the polycarbonate. Outgoing air downwards between the sheets with the opportunity for any condensation to trickle down and dribble out of the bottom of the exchanger. The whole lot would be boxed up in bolted-together painted ply with plastic spacers of some sort between the sheets.

In some ways it would be easier from a drainage and airflow point of view to have the outgoing air go through the channels in the polycarbonate but if there are any problems with condensation causing black slime then cleaning the outside of the sheets would be a lot easier.

Overall the house will be supervised by a generic low-power Linux box with code in JavaScript under Node.js to start with. Different sensors and actuators will often have their own processors doing low-level control - JeeNodes and the like are favourites at the moment but I'm deliberately being a bit technology agnostic, not going down the path of trying to make everything look like 1-wire or whatever.

Absolutely agree with joe90 - the overall supervisor code has to think reasonably strategically about when to operate different functions of the house. E.g., it should tolerate the humidity rising a bit over a cold night, particularly if the batteries are a bit low, if it knows that the following day is forecast to be bright giving warmer air to dry the house out and sunshine or wind to power the fans. In other words, control should be based not just on current conditions but also on their projection ahead.

Posted By: Andrew_Doran.. add to that, the wired remote control for the Sentinel range (I have one) uses that interface and it does quite a bit.

However, according to the Installation and Commissioning Manual the control program has two BMS modes: On for BMS (default) or Off for Wired Remote Control, automatically set up by BMS signal or Wired Remote Control when either is plugged into BMS RJ11 socket so there might be something a little more subtle going on.

I have no reason to disbelieve the document that Vent-Axia supplied. So "Sentinel Stop" and "Sentinel Start" are the only officially supported options in BMS On mode. It did provide details of Data that could be output to a BMS so my guess would be that if a large customer wanted a modified interface that could be supported by a software change.

Must admit I was rather disappointed that A full range of two-way digital signals are available through the RJ11 connector on the control board turns out to be a couple of one-way inputs! A simple, read only interface to the internal data would be low risk and offer much greater flexibility. Even the ability to input external temperature, humidity or CO2 values may not have required much additional work.

Posted By: dgmartinLike john_connett, I have Ethernet nearby and (as it happens) a raspberrypi monitoring the temperature (via Ethernet connected DS1820 probes) all round the house, so perhaps the Rpi could do a bit more work controlling the MVHR too!

That could be a serious long-term project and would involve ripping the Ventaxia electronics out; providing new drivers for the fans etc, and creating a new user interface, with some sort of (much better) remote management and monitoring.

As far as I can tell, Vent-Axia have made reasonable design decisions with their control board and have included scope for future expansion via the RJ11/BMS and the NET connector (currently unused?). They have sent prompt and helpful responses to my e-mail enquiries. Building Management Systems and The Internet of Things may be close to the edge of their normal product range but they appear to be moving in the right direction with an understandable degree of caution. It may be that a software upgrade (or control board upgrade) will provide the enhanced control options that some of us would like to see. One way for a company to estimate demand for enhancements is from customer requests. If you would like something then ask for it! They might say no or they might see an opportunity to extend their product range into a wider market.

It would be interesting to know what control options are provided by other domestic ventilation suppliers. I suspect that the air conditioning manufacturers probably provide better BMS interfaces because they have probably been doing longer. Same might go for the likes of Vaillant and Viessmann that provide a wide range of heating and ventilation products although often using their own proprietary interconnections.

I have a vague feeling that I read about an extraction ventilation system in Green Building Magazine that lead all the ducting back to a single box with butterfly valves where each valve entered the box. Pretty sure it had a CO2 sensor (or sensors). Unfortunately, can't remember the name. Sounded interesting but lots of moving mechanical parts are not ideal for reliability ...

My guess is that most MVHR systems with RH sensors (internal or external) would boost flow based on the highest RH reading from any sensor which seems like a sensible compromise. A manual boost is useful to remove smells when the RH is within bounds. In my case, I can do that by turning on the cooker hood fan which has a volt-free connection to the MVHR boost.

Pretty sure this isn't the one that was in the magazine but it has similar features:

http://export.renson.eu/Mechanical-extract-ventilation-unit-Healthbox-II_4.html

However, this is demand controlled extract ventilation so doesn't have heat recovery.

Oh yes I'd say they were worth the money. As to actual costs try the web and look for units like "Ventaxia Sentinel Kinetic Plus". A quick looks shows prices around £1500 but fitting and lots of duct, terminations, roof vents etc will add considerably more. Of course there are other units available too. Our discussion above was along the lines of how manufacturers could do more to make the units better controllable and flexable. But in general the units do a job reasonably well.

I have one of these units as well. My idea was to use one of the proportional inputs, driven from a arduino or Rpi (have several of both). From the commissioning manual it states.

When the input signal is below the ‘Normal Limit’, the
unit runs at low / normal airflow. When the signal is
above the ‘Boost Limit’, the unit runs at boost airflow.
Between these limits the unit runs at a proportional
airflow.

So I was just going to set normal very low, Boost very high, and control the range between myself. Has anyone done this yet, I little confused as to exactly to produce signal. (e.g. how much effort I need to put into smoothing a PWM output)?

Just been looking for Pi hats with 0-10V output. Rather surprised to only find one thing of that sort and that's quite a complicated beast for about €140. Anybody know of anything I've missed?

That's an ADC (analog-to-digital converter) for input. It could do 0-10V input easily enough with just a voltage divider (two resistors).

But I was asking about a 0-10V output to drive one of the inputs of the Vent-Axia. There are plenty of circuits around which will take the output of a DAC over 0-3.3 or 0-5V and push it up to a 0-10V output but it would be much nicer to have a pre-made hat which will just do it.

Yes, would either need an external higher voltage (12 V probably) supply or an onboard boost converter. Maybe there are chips to do it directly (i.e, DAC and boost converter combined) similar to the MAX232 which generates ±7.5 volts from a single-ended 5 V TTL supply for RS-232 outputs (with the aid of a few external capacitors).

Whatever, a DAC, op-amp and boost converter for this on a hat seems so obvious I'm a tad surprised there doesn't seem to be one.