JNIOR application - heating

[Carsten Kurz]

Okay, as we're all in a 'hold' situation, why not take this http://www.film-tech.com/vbb/forum/m...=1764#post1764 to here:

Since a while, I was wondering wether I should employ a smarter method for our auditorium heating. It is a decent room, about 14m wide, 24m deep, and about 6m high. It's a free standing historic building, so, little insulation, loads of heat loss through walls and ceiling/roof. Over the years, we employed a few methods and that essentially brought the heating cost down to around 50% (which is great, given that it didn't cost us much). However, I think we could do a little more. The auditorium heating is using hot air. The inlet can be switched from external to circulating air. During our heating season, we switch to circulating air, as then the energy is reused (heating up air in the range of 18-20 °C costs less energy than heating up external fresh air from e.g. around 0 °C). As the auditorium has such a huge air volume, as we only have one show per day, and the building is anything but air-tight, we usually don't have issues with air quality. Only when we are sold out (about 350-400 people) we are switching to external to improve air exchange. We don't employ explicit HVAC. During summer time, we can switch on just the blower of the heating system, and while being switched to external inlet, this also brings in fresh or cooler air from the outside. This, currently, is all manual.

Now, the time to heat up our huge auditorium is comparably small, as the heating system blows in a huge amount of warm air in comparably small time - we usually switch the heating on around 1hr before admission starts (during times with very low outside temperatures, this needs to be increased to 1.5hrs, sometimes 2hrs). This is done using a standard wall mount thermostat/week timer combo set to 1hr before scheduled admission time. This is adjusted manually, so when we start a long movie half an hour earlier, I have to adjust that timer on site. Similarly, when we have extra shows e.g. in the morning, I need to add extra timer points beforehand on site. I am not living near the cinema, I am not on site every day, and I'd hate to drive there an extra 1-2 hrs earlier in the morning or evening just to switch on the heating in time. Our general staff is not good in handling that clumsy current timer 'interface'. They could easily ruin important settings/schedules, leaving the heating off during show times, and when they usually arrive at the cinema for their duties, it is usually too late to enable the heating manually in order to achieve a decent room temperature until admission or even show starts.

The heat-up/preheat times have been found empirically using a range of temperature loggers some years ago. We found a more or less linear relation between time and temperature rise - around 1 °C per 10min of heating. The lowest I let the auditorium temperature drop in harsh winters during night and most of the day is to around 7-8°C (that is the lowest safety temperature our thermostat allows, and I also think this is still safe for building and interior. Our utility rooms use a separate heating and never drop that much). So, from these numbers, we can see that, to bring the auditorium to around 20 °C, we need a maximum of around 120min heating time. This has been verified in daily use over the recent years. The auditorium air temperature is measured using an external temperature probe attached the left side wall, half way into the auditorium. There is probably a bit of a bias there being so close to the external wall, but, this is usually easy to compensate by using a fixed offset. During the movie, the heating switches off and on based on room temperature and hovers around 20 °C. So, with a huge attendance, we use less energy, as their body-heat adds to the heating, means, we never over-heat.

However, me adjusting the timer pre-heat time of cause doesn't happen on a daily basis, I assume a general mid-term average temperature and set the timer accordingly, but, by a very rough estimate. So, during early autumn, the preheat time may be only 30mins, while in a cold January, I may set it to 2hrs. The Off-Time in the timer is usually set to straight times roughly based on the show length, but, not very precisely. E.g. it may be set to 22:00 for most shows. Only when we play an exceptionally short or long film this is adjusted for it's run time. We notice that the auditorium temperature drops 'noticeably' within around 15min after switch off, and we don't want people to start quivering already during the end of a movie.

I think this can be impoved by setting the preheat time automatically based on the auditorium temperature, and that is when a JNIOR comes into play.

- measure room temperature and switch on heating automatically based on a pre-heat time calculated from the room temperature and scheduled admission time, so, if the room is e.g. 12 °C, start heating 80min before admission time
- receive notification from the cinema server about show end and switch off automaticallly. This could e.g. be based on credit offset.
- maintain the existing manual timer and thermostat as a backup, so that, if anything goes wrong, it switches on heating a minimum of 30min before admission, and at a fixed safety time after a movie ends. Failsafe is important here. If necessary, I need to be able to tell general staff on the phone how to bypass the system if it fails for any reason. This also has implications on the way we wire/choose relay contacts break/make wise, e.g. a broken/hungup/disabled JNIOR or software function should not block the failsafe timer from working.

My idea would be to enter show times e.g. in an electronic calender, as currently, we don't have a TMS or POS.

Some luxury items could be added later - e.g. alarm if the room temperature doesn't rise even if it should (problem with the heating system not firing up). Notify in early autumn to enable the heating if the average temperature decreases (our heating is shut down completely from late spring to early autumn).

Now, for a short time a few years back, I actually had connected a JNIOR 3xx to the heating and was able to turn on remotely. However, we had a problematic relation with our landlord back then, and he did not allow us to mess with 'his' heating system. Now, the landlord has changed, and I'd like to pursue this task. The turn-on element for the heating is a simple 230VAC relay accessible from the projection room. This is in series with the thermostat, so, IF timer AND thermostat condition is met, heating is switched on. The current thermostat uses a 20mA wired probe in the auditorium. I would love to have a wireless probe, though, we could wire one, however, we would probably need 20-30m of cable run. Not sure yet which position of the probe in the auditorium would be best.


So, Bruce, or anyone...?

- Carsten

[Bruce Cloutier]

Well... I was working on label artwork. I am so damn DIY. We just invested (before this pandemic) in UV printing equipment so INTEG can produce its own quality product labels. We've got an updated look for the JNIOR. I was hoping to debut that at CinemaCon but...

So I was VP of Engineering for American Auto-Matix around 2000 during the OJ Simpson trial and 9/11 days. AAM is an HVAC controls company. They actually grew out of a Carnegie-Tech (CMU) and were the very first to apply microprocessors and networking to HVAC (c1970). Before that there were the vacuum based systems some of us old-timers will remember. So this topic...

The JNIOR is very appropriate for HVAC control applications. One issue with that market is that they have bought into the BACnet protocol. That is supposed to help systems inter-operate where before each had its own proprietary control protocols (like Cinema). So without a customer or two hanging an HVAC carrot in front of us, we/I haven't done the BACnet implementation. But, that really only applies to big commercial systems and most of us have some more mundane heating and cooling solutions in place. So we need to control with relay closures.

INTEG has two separate and practically residential style furnaces and roof-top air conditioning units in use. The thermostats fight each other and do not automatically switch between heating and cooling. So during the day we manage those manually, constantly. I have been swearing up and down that I would put JNIORs there to do that. But, our priority is you guys.

There are 2 additional things Carsten that you might consider. First there is Cold Soak. Materials absorb heat at different rates. So when you bring the room temperature up starting an hour before the show, patrons in seats might still be uncomfortable as their backsides have to deal with colder cushions and arm rests. It takes a long time for everything to come to equilibrium.

Also, Relative Humidity plays a big role in comfort. When you let a space drop in temperature you drive moisture out of the air. You might actually have some condensation issues that could lead to rust on metal surfaces or mineral deposits on lenses. Then when you heat a space the relative humidity is low and limits heat transfer with your skin. As a result you have to heat the space much more than necessary to make sure everyone is comfortable.

I have a JNIOR in my home (go figure) and its job is to manage a steam humidifier that I added to deal with this humidity issue. The standard evaporator humidifier that they install on furnaces just doesn't do it. So during the cold weather when the humidity in the house gets way too low the JNIOR kicks in the steam humidifier to assist the standard one. Now that uses amps and costs electricity. So its use is limited. During the night you can also hear the water boiling in it. That can be annoying when you are trying to get to sleep. So the system disables it from 10PM until 8AM. The JNIOR does that really nicely. I can share all of the details. There might be something in our Knowledge Base (jnior.com). I can't remember what I've written and what not.

So there is that question as to how much noise your air handlers make. Also, the efficiency impact of dirty filters which you can detect.

But, yeah, it would be a simple matter to use temperature sensors, wire JNIOR relays and to develop some HVAC control programming. You have the ability to do that on your own. Um, try to find a Series 4 though. You can do it with the 310 era product but it is so much easier now.

Put the JNIOR on a network and wire relays instead of manual switches and you can remotely control using the web pages and not have to write code. Wire the switches into inputs for manual overrides.

[Carsten Kurz]

Hi Bruce - those temploggers we used to analyse the heating scheme also have humidity sensors, and I kept a close eye on their curves as well. In general, and as we don't have much tempstorage capacity in the auditorium, I think we are doing well. Also, I am not after precisly hitting 20 °C when the first patron enters - if we need 15min more to keep them comfortable, no problem to cater for that.

There are some non-technical factors there as well - those people entering the auditorium come from a cold outside. Those early ones have usually been waiting at our front door in the cold. So, even if the auditorium incl. seats have not already warmed up enough, they will probably still feel comfortable and then acclimatize. Also, at least around here, people dress sensibly during these periods, and so will usually have some clothing that compensates a bit.

But yes, it is very important to keep humidity and it's balancing with fast rising temperatures in mind. As I said, our auditorium generally has a lot of air exchange due to it being an old building, it's a sieve.
If we would need to heat it 24/7, that would be a nightmare. But, with that amount of air exchange during the heating season, I don't see an issue with too much humidity or condensation building up - it is quickly vented out.

It certainly is an issue when we are sold out and towards the end of a show with all the peoples breathing, possibly wet clothes, etc., and then switching off the heating. But that happens quite rarely.
Also, another luxury addition would certainly be a humidity and/or air quality sensor.

I do have a JNIOR 4xx as well. My main issues so far would to be how to enter show times easily (also possibly for my not so technical colleagues), how to connect a suitable thermoprobe, and then how to calculate the pre-heat time from these two parameters. Also, to initially find my way through the many ways a JNIOR would allow me to address this project. I did a lot of reading through your docs when I was playing with the 3xx back then, but lost a bit of that in the meantime. But again, I have some spare time now.

I understand I wrote a longish text up there, but, after having set all conditions, maybe we can brake it down into simpler tasks that center more around the JNIOR.
E.g. how can I input and keep a short- to midterm schedule of show times in a JNIOR.

- Carsten

[Bruce Cloutier]

Kevin is really our applications expert. You can hit him up with questions using our support [at] integpg.com address. We can address some of this with off-the-shelf applications like Task Manager. He and Tony are presently working on Tasker which is a less cinema-esque scheduling package. JNIORs can also talk to each other easily. So there is the possibility of one telling the other to attend to the environment as a show is pending. That would save entering and maintaining a schedule in multiple places.

Kevin can offer ideas and tricks to experiment with while the scope of what you want to do solidifies. I'll point him here to this topic but, unless others encourage us to keep it going here, you guys can probably play with things offline.

Our Sensor Port can handle 1-wire devices. We have a temperature probe and also a temperature/humidity probe. Some are integrated into the OS and other devices can be read and written without too much trouble. We can investigate some things.

By the way, We haven't jumped on the bandwagon and issued any COVID-19 statement but we are shutdown by order of the state. Everyone is working remotely. Production has stopped. We'll take orders but won't be generally shipping until we can reopen.

[Carsten Kurz]

I know that 1-wire can have wire runs of 50-100m when properly terminated. Did you ever test this with your sensors?

[Bruce Cloutier]

We have tested to over 100m but I would not say that we did so scientifically. It is not just the distance that matters but the quality of the wire and the environment it encounters. It also matters as to what you have out there sharing the network. We operate that network at normal speed (~9,600 baud) and cannot utilize overdrive. We use a 5V line and do not derive power from the 1-wire. Our modules communicate over that bus and they are not strictly 1-wire. Some others have tried to talk to our expansion modules using some other 1-wire driver with no real success.

We can, however, communicate under reasonable conditions with any 1-wire sensors that can be powered separately. That is the case with our temperature and environment sensors that use the 5V supply. Also the current to that 5V line is limited. It is just not designed for a large sensor array.

I think that we have run something like 12 of our LED dimmers on one port without difficulty. These were tethered close by. Our LED dimmers power the LEDs from a separate DC high current supply and so the Sensor Port only powers the MSP430 controlling each dimmer. Those are really low-power and isolated from the serious side of the dimmer.

But you are welcome to see what you can do. The wiring is published and the Java classes that you need to communicate at a low-level with any random sensor there. There are examples. Our modules and even the temperature sensor are also integrated with JANOS so there are simple high-level ways to communicate with those. They show up on the External I/O tab on the default web page.

The JNIOR on my furnace has 4 temperature sensors connected via telephone splitters (RJ12 6p6c). They measure delta temperature on the heat exchanger and on the ground water lines. I have geothermal for heating and cooling. There is a high-efficiency gas furnace for backup heating. Kevin has a prototype version of Google Blockly we call Snap running that captures and logs temperature readings. On the web side there is an application called Grapher that presents plots. I can look minute to minute back months. I also have 20KW of solar (http://cloutieronline.net) although the JNIOR is not employed there.

I think it was you, Carsten, who said that the JNIOR is underutilized. You all already have the JNIORs and all of this application fun is free of charge. You just have to be creative. Uh... did I suggest Series 4.

[Frank Cox]

I'm probably showing my ignorance here but why would you use or require a 9600 baud connection for a control application? It seems to me that 300 or maybe 2400 baud might get you more range and greater accuracy.

When all you're doing is flipping a switch, what does a higher speed get you?

[Bruce Cloutier]

Hi Frank,

Well in this case the 1-wire designer elected to use a bit rate of 9,600 baud (for some reason that no one at Dallas Semiconductor ever documented I don't think) and that, for us, is still somewhat slow. Imagine if you are trying to read a couple of thousand temperature sensors under tiles on the Space Shuttle (not a JNIOR application) where each reading takes bytes of addressing and error checking in addition to the accurate digital temperature reading. So the 1-wire specifies an overdrive mode that uses a higher bit rate. Even that is limiting.

But it is a fair question. In our market here we are of course more concerned about controlling something than monitoring it. MODBUS is a great protocol (Master-Slave) for controlling but not for monitoring. It is similar in some ways. To control you just send a bit as you say. But to detect a change in a reading you have poll for that reading over and over. 99.9% of the responses are the same. But then one is different. And you would like to know about that change ASAP so you ask as fast as you can. So polling takes bandwidth and a bit rate of 9,600 baud (typically 1 byte/character/digit per millisecond) with protocol overhead, you would be lucky to make 50 readings per second or read 50 sensors once per second. Again, if you are monitoring the temperature of a room, no big deal. Something a little more elaborate like the Space Shuttle, well...

So the JNIOR commanding an external 4ROUT relay module is no challenge. But using an external 10V module to accurately monitor the RPMs of a wind turbine to be used to control bade pitch and power transfer, would be better at even higher baud rates (an actual JNIOR application).

I should have used an isolated RS-485 network for this port on the JNIOR, but guys before me were into this 1-wire stuff. We've been debating that change for years.

A reasonable question though, Frank.

[Carsten Kurz]

Okay, so I guess I'll have a go. I knew about the temperature sensors, but wasn't fully aware about them being digital and able to work over a decent distance. That makes a JNIOR the perfect choice. Our heating season will end soon (probably before our closure ends), so time and options to play with this until autumn.

1-wire certainly has advantages over RS-485 cost and options wise. I mean, RS232 and RS-485 is available on all or some JNIORs anyway.

[Bruce Cloutier]

1-wire certainly has advantages over RS-485 cost and options wise. I mean, RS232 and RS-485 is available on all or some JNIORs anyway.

I think the AUX port on the JNIOR3 all are RS-485 capable but the firmware inside less so. No body used it for a decade. So it was a wasted cost. With JNIOR4 we left RS-485 capability in the 410 but not the others. We've picked up a few RS-485 users in the recent years and the firmware makes it quite usable now. DMX lighting is basically RS-485. It is at an odd baudrate of 250K and the format uses the serial break condition. Anyway, we have the 412DMX which supports a DMX universe as it is called. We're not pushing that model though (do we push anything?) and it might go away if there isn't any real interest in it. It hasn't proven to be a valued addition to the line yet.

RS-485 vs. 1-wire in the expansion modules would be a wash cost wise. I have thought about making the modules compatible with both. That might make the modules useful without the JNIOR. But on the JNIOR side the bigger hurdle is the legacy issue. Both formats there would be expensive to support together.

The real advantage though in our using RS-485 is that the communications could be at a tremendous baudrate and bi-directional. A peer network eliminates the polling requirement. If you are monitoring you don't have to repeatedly read the data. The module can advise you when its inputs change.

We've had requests for applications that would sample the 10V signals 1000s of times per second performing some AI analysis. We can't support that. Our modules average and you can read one 10 times per second. Half that if you have two modules. In this market there is more interest in the 10V outputs as dimming reference levels. Here the fastest you can update that is 10 times per second and that is too slow for any kind of smooth dimming. So the modules include the ability for smooth transitioning among other features.

Of course for the right application I can make a JNIOR with the right I/O internal and then it could kick ass.

[Carsten Kurz]

Anyway, we have the 412DMX which supports a DMX universe as it is called. We're not pushing that model though (do we push anything?) and it might go away if there isn't any real interest in it. It hasn't proven to be a valued addition to the line yet.

I guess that, unfortunately, it came out a little bit too late after most of the digital rollout was over. These things are not picked up too often once they have been thrown in place. As I said and as you quoted me - in many cinema installs, the potential of the JNIORs is not nearly used. In most installations I saw, they are used as rather dumb network relays in their factory config with a Doremi driver to work a curtain or dimmer contact. I guess many integrators did not find it worth the time to dig deeper into them - also because they were very busy at the time and needed quick and simple solutions.
This is probably very different from the industrial automation business where it is quite common to tweak and streamline their use to the max.
One major issue is that in many retrofit installs, it was very easy to access auditorium automation through relays based on either existing film based automation or button interfaces. Not, Heating/HVAC is a completely separate business which cinema integrators never tough, at least not in retrofit jobs.
In new installs, nowadays someone will usually push through interfaces to common building automation standards already in the planning phase, but that is rarely done with a JNIOR.

- Carsten