Boiler Computerization

Reply

  #1  
Old 08-30-13, 05:47 PM
Member
Thread Starter
Join Date: Oct 2008
Location: USA
Posts: 2,880
Received 5 Votes on 5 Posts
Boiler Computerization

I like computers. I designed and built, from scratch, a microcontroller to manage my electrical demand charge by cycling my air conditioner. I've programmed in assembly language, Fortran, and Basic. I'm not a complete luddite.

But: hot-water boilers, as simple as they are or should be, are being majorly computerized - error codes, etc. Why? The trend is certainly not DIY friendly, at least to me. It used to be that troubleshooting could be performed with an multi-meter. Now, it seems a factory-trained technician, with perhaps a proprietary test device, not a plumber, is required.

I have a new Ford, with their "Sync System," designed by Microsoft. The user manual for Sync is 72 pages long. I got my cell phone to connect through the radio and Bluetooth, but beyond that, I quit - I don't have the time or patience. Even the cell-phone connection, with audio commands, is troublesome.
 
Sponsored Links
  #2  
Old 08-30-13, 06:16 PM
Member
Thread Starter
Join Date: Oct 2008
Location: USA
Posts: 2,880
Received 5 Votes on 5 Posts
Additional Rant

With my Ford Fusion, I can't even rotate the tires myself. Well, I can, physically, but then the computer needs to be reprogrammed, requiring a special tool and a $12 charge, so it "knows" that the left-rear tire is now on the left-front - otherwise the computerized tire-pressure monitoring system will be confused. The end of DIY?
 
  #3  
Old 08-30-13, 06:32 PM
Member
Join Date: Mar 2006
Location: Wet side of Washington state.
Posts: 18,482
Received 25 Votes on 19 Posts
Why the microprocessors on boilers? Because they can. Everything mass manufactured in China for next to no cost makes it definitely less expensive than discrete pressure and temperature gauges and switches. Add in the burner management system and it is way cheaper than discrete components. Repairs are simple, just replace the board no matter what the problem. If that doesn't fix it then replace the sensors. Perfect for the non-thinking technician.

Plus it has the WOW! factor.
 
  #4  
Old 08-30-13, 07:16 PM
Member
Join Date: Nov 2011
Location: USA
Posts: 408
Received 1 Vote on 1 Post
Regarding heating and cooling, the holy grail these days is "energy efficiency", and to me that seems to be a key driver of the ever-increasing complexity of HVAC controls. Whether they actually deliver on that promise is another issue.

Regarding automobile entertainment-delivery systems, the main focus today is "connectivity", and likewise, that is why you see 72 pages in your Focus OM dedicated to the Sync System.
 
  #5  
Old 08-31-13, 08:23 AM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 2 Votes on 2 Posts
It used to be that "form followed function", but not any longer. The advent of the micro-controller means that pretty much any form can have any function.

If you could pull your Fusion into your basement and apply some programming changes, you could use it to heat your home.

Your telephone is a camera, audio device, notepad, and electric razor. Well, it probably COULD be a razor. I still can't get mine to toast bread though...

Bottom line; It's all about economics.
 
  #6  
Old 08-31-13, 09:56 AM
Member
Join Date: Mar 2006
Location: Wet side of Washington state.
Posts: 18,482
Received 25 Votes on 19 Posts
The microprocessor costs the manufacturer maybe $25. The sensors maybe $25 each, maybe less depending on what is monitored. Far less expensive than the pressure and temperature switches of years past. Once the microprocessor is in place the amount of money invested to program it to do a dozen more "gee whiz, ain't that snazzy" functions is negligible.

So, yeah, it IS all about the money. What blows me away is that the manufacturer uses the electronics to save money yet often charges a premium over more conventional controls. Just more of that "WOW" factor.
 
  #7  
Old 08-31-13, 12:06 PM
Member
Join Date: Nov 2011
Location: USA
Posts: 408
Received 1 Vote on 1 Post
I certainly see the point about micro-processers and their relationship to the almighty bottom line.

But you can't ignore the fact that microprocessors also allow for the installer/operator to have more control over a system, which presumably leads to better energy efficiency.

Let's take the example of ODR and outdoor temperature measurement and feedback. The purpose of ODR is to make a heating system more efficient and responsive. So how does one accomplish this - at least in a simple residential setting - with older pressure/temperature gauges and switches? How would one go about rigging a system that can (almost) instantaneously measure outdoor temps. and send that data to another control, which control then raises or lowers the high limit (and maybe even adjusts the differential) accordingly, without any need for human interaction?
 
  #8  
Old 08-31-13, 01:38 PM
Member
Join Date: Mar 2006
Location: Wet side of Washington state.
Posts: 18,482
Received 25 Votes on 19 Posts
Absolutely nothing new about outside temperature reset; they had mechanical ODR systems for residential systems some seventy years ago. They used linked dual thermostats with the capillary tubes sometimes as long as fifty feet. One bulb located in the boiler and the other located outside, generally on the north side of the building or shaded from direct sun. Some of them had adjustable "slope" or reset ranges although admittedly nothing like the adjustment available via microprocessor units. By the mid 1950s the outside bulb and capillary tube had been replaced by an electronic sensor, usually a thermistor because they were cheaper than the full mechanical units. Next step was electrical sensor in the boiler as well.

These older systems worked well in their day and the ones that still exist I suspect are still working as well today as when first installed. Now I will grant you that the modern full electronic systems are a bit more accurate and faster operating but in the real world those two items are probably of little concern since the relationship of boiler water temperature to outside air temperature for maintaining comfort within the house is pretty well fixed for any particular installation and simply does not require either precise measurement or high speed operation.

Better than ODR is a system that monitors the inside temperatures and adjusts the boiler temperature and firing rate (modulating burners) to maintain more accurate and consistent indoor temperatures. If you remember back a couple or three years there was quite a discussion with a man that had devised such a system for an apartment house or college dormitory. A member who has been absent for quite some time (and was instrumental in my joining) was using a variation of an indoor reset system and he had nothing but praise for it. As I recall he said it maintained his indoor air temperature within 1/2 degree throughout something like a forty degree outdoor change.

Don't misunderstand me, I LIKE digital control systems. That is one the primary things I was doing a few years before my retirement. But never lose sight that the reason for the manufacturers to go digital was to SAVE MONEY, not improve the product. The improvement in the product was just an added benefit.
 
  #9  
Old 08-31-13, 03:35 PM
Member
Thread Starter
Join Date: Oct 2008
Location: USA
Posts: 2,880
Received 5 Votes on 5 Posts
ODR reset is probably a good application of digital controls. But...I've been waiting for somebody here to bring up programmable thermostats. I tried two of them, and hated both. The wife hated the programmable 'stats and came close to hating me for trying them.
 
  #10  
Old 08-31-13, 07:13 PM
Member
Join Date: Nov 2011
Location: USA
Posts: 408
Received 1 Vote on 1 Post
Absolutely nothing new about outside temperature reset; they had mechanical ODR systems for residential systems some seventy years ago. They used linked dual thermostats with the capillary tubes sometimes as long as fifty feet. One bulb located in the boiler and the other located outside, generally on the north side of the building or shaded from direct sun. Some of them had adjustable "slope" or reset ranges although admittedly nothing like the adjustment available via microprocessor units.
I don't understand how this form of ODR would work. Isn't a t-stat just a simple switch? How can a switch be used to generate the kind of data that would allow the boiler to adjust its limits on the fly? What criteria is used to "set" the t-stat?

 
  #11  
Old 09-01-13, 12:45 AM
Member
Join Date: Mar 2006
Location: Wet side of Washington state.
Posts: 18,482
Received 25 Votes on 19 Posts
I didn't realize I was going to have to explain how the mechanical systems work but I'll give it a try.

They utilized a thermal-hydraulic system with the two systems working on both sides of a dingle diaphragm. The sensing bulb on the outside of the building would respond to the outside air temperatures moving the diaphragm one direction on an increase of temperature and the opposite direction on a decrease of temperature. It is actually TWO diaphragms with a mechanical link between. This is exactly how mechanical aquastats work.

Now add an additional sensing bulb on the opposite side of the diaphragm but this bulb is located in the boiler water. The diaphragm itself is mechanically linked to a spring and microswitch. The spring is the opposing force from the switch and is adjusted to allow for different temperatures in the boiler. There is also a sliding link adjustment between the two diaphragms and this is on an angle. Adjusting the angle of this link changes the response of the switch relative to the difference between the outside and boiler temperatures. Because the link is on an angle the term is slope.


Programmable thermostats...

The first programmable thermostats were nothing more than dual thermostats wired in parallel with a clock-driven single-pole, double-throw switch to select between day and night temperature settings. They would allow for a night set-back temperature and then a rise back to "normal" temperature sometime before the family woke up in the morning. Having analog clocks and being totally mechanical they could not be set all that closely and were completely at the mercy of power failures except for the models that had spring-wound clocks and those suffered from all the same problems as any spring-wound clock.

As time has passed the programmable thermostats have improved. Electronics, especially integrated circuits have made huge inroads in this area. When first introduced the Honeywell Chronotherm (the first of the mechanical programmable thermostats) had a price above $50 when a non-programmable thermostat sold for less than $10. You had to save a significant amount of money to justify a thermostat that cost more than ten time what a regular thermostat cost. Today you can buy an electronic programmable thermostat for maybe double what a non-programmable sells for.

The early electronic programmables were not very smart. They allowed the same two temperature settings as did the original mechanical models and the minimum timing period was usually 30 minutes. That meant that you could have the set-back begin or end at either the hour or the half-hour and not at ten minutes after or twenty minutes before the hour unless you purposely set the clock incorrectly. They were a single program for every day of the week as well, just like the old mechanical thermostat. From this beginning the manufacturers added features like fifteen minute time settings, 5/2 day programming, 5/1/1 day programming and then true seven day programming. To my knowledge (I don't keep up on these things since retirement ) they have NOT changed the fifteen minute timing period. One important thing they HAVE improved upon is a learning ability that allows the thermostat to learn the characteristics of the heating system and then adjust the programmed start (morning pick-up) time for maximum system efficiency. With the earlier thermostats the user programmed the time he/she wanted the heating system to begin the day (or morning) temperature setting. Since the thermostat had no "understanding" of how quickly or slowly the house would achieve the desired temperature it was necessary for the user to set a time far enough before the alarm clock to ensure a warm(er) house when they awoke. In milder nights the heat may not come on at all and the temperature may not have fallen all that far but the thermostat would still wake the system at four AM to ensure the house was at the programmed temperature by seven AM.

The latest programmable thermostats however use the learning feature and will delay the start of the system to the last minute before necessary to reach the desired temperature at the desired time. This has made user programming far simpler in that the user no longer has to make some arbitrary adjustment in the start time but instead just programs the desired temperature and when that temperature is wanted. For example, if a person wants the house "up to temperature" at 7AM they set the thermostat for 7 AM and the thermostat will calculate when the heating system must be started in order to have the house at the desired temperature at 7 AM. If it is cold outside and the temperature of the house fell to 62 overnight the thermostat will start the heating system at an earlier time than if the outside temperature is warmer and the house temperature only fell to 66 degrees.

The most unfortunate thing in regard to the programmable thermostats of today is they are STILL locked into that silly fifteen minute timing window for temperature changes. That and the FACT that the programming interface in most is absolutely archaic in that each day in many cases needs to be separately programmed and it is a huge pain in the rear to either review the entire program or to change just one aspect. I guess that is one of the reasons for the newer units to offer WiFi programming but in my mind a simple USB interface along with a spreadsheet type of programming interface would be far superior.

Does this help?
 
  #12  
Old 09-01-13, 01:28 PM
Member
Join Date: Jan 2008
Location: Southeastern Pennsylvania
Posts: 2,947
Received 0 Votes on 0 Posts
gilmorrie -

ODR reset is probably a good application of digital controls. But...I've been waiting for somebody here to bring up programmable thermostats. I tried two of them, and hated both. The wife hated the programmable 'stats and came close to hating me for trying them.
gilmorrie I agree with you and your wife. Some of the features in the programmable thermostats are really questionable IMHO and they can get in the way adding confusion and unneeded complexity. I think they go way overboard with the “scheduling” choices they offer. Who needs it? Does anyone really use all that? lol Yet you have to understand it all to make sure the thermostat will give you a simple constant temperature over time. That’s aggravating - to me anyway! lol

I’m a retired software guy. Been away from it for years. But I know that once a microprocessor controlled system is developed, the temptation to add features, even though they may be near superfluous, is great. The reason is that the vast majority of effort goes into the significant initial development and integration of the hardware and software to produce the minimum core requirement. You probably know this from developing your own system. (We actually used to have big parties at the core-completion milestone.)

After that, for some (but only some) features, it just becomes a matter of adding a relatively small number of computer instructions to be able to add that feature. Everything else is already in place and is working. There is however, some additional cost to add more features such as cost of software integration, testing ,user manual updates, etc. But relative to the initial development cost for the system those costs are small, although sometimes (lol)…adding that “one more” function implemented by just a “few more instructions” may come back to haunt you when you later find out it actually causes problems. – I’ve seen that !

I guess, unfortunately, the dollars and cents guys with the MBA’s rule. That’s the way it seems to me anyway. They feel forced to add a myriad of features for their system to be competitive. If your manual is thinner than the other guy’s manual, your system will appear too simple, i.e., deficient in some way. So it’s off to the races and we get these complex manuals with a majority of features that almost no one uses. IMHO. But I do see a paradox in what I am saying. If we consumers demanded simplicity then you would think the systems would move towards only the absolute minimum complexity required?

I worked on mostly custom systems where the end user tells you exactly what to include in the system and so you don’t get those bloated monstrosities. But even there, the customer after having spent millions on a system, will be tempted to add more and more features when he sees that the relative cost for an enhancement is peanuts compared to his total outlay. Guess psychology is a big part of that.

I do think however that the use of a microprocessor base allows very useful and significant additional functions to be added relatively cheaply once the initial system is built.

For example, we built a system once to replace a manual system that transmitted extremely critical messages via earth orbit satellites. Later, we were then able to add a function, just by adding a relative small number of microprocessor instructions, that would compare exactly what was transmitted to the satellite (the uplink) with what was received back on the earth (the downlink). If the 2 message were identical bit-for-bit, we would then turn on a visual indicator (a light already on the system).

Thus the operators of the system, just by looking at a light, could be very confident that the message they sent was exactly what was re-transmitted by the satellite back to earth. That same comparison could in fact be done manually, but it was time consuming, not in real time, and prone to error. And then with just a few more instructions, in the right places, we were also able to make the light blink ON-OFF while the message was being transmitted up to the satellite. The result was, that after the user released his message he would then see a light per satellite blinking ON-OFF while the message was being transmitted uplink to that satellite, and then the lights would stop blinking and turn ON solid, one by one, as the system confirmed that the downlink copy exactly matched the uplink copy for a given satellite.

That is just one example of a extremely significant enhancement done relatively cheaply after the microprocessor based system was in place. I think trying to implement something like that without the microprocessor would be unbelievably complex, cost prohibitive, and fraught with problems.

IMHO, in general, micro based systems replacing older systems allow significant enhancements to the replaced systems and are a boon to mankind (lol), but it’s just that sometimes the marketing gurus throw in the kitchen sink so that their system looks bigger and better. Maybe I have what is known as “an intuitive grasp of the obvious” lol.

OK I’m done, too much time on my hands.lol Did I set the forum record for ramble? lol

p.s. I do like the fact that the programmable t’stat displays Current Temp, Setpoint, and especially “Heat On”. The rest of it, especially that scheduling stuff, is all fluff! lol
 
  #13  
Old 09-01-13, 03:03 PM
Member
Join Date: Nov 2011
Location: USA
Posts: 408
Received 1 Vote on 1 Post
Does this help?
Yes it does. Thanks for taking the time to explain.
 
  #14  
Old 09-01-13, 03:26 PM
Member
Join Date: Nov 2011
Location: USA
Posts: 408
Received 1 Vote on 1 Post
That is just one example of a extremely significant enhancement done relatively cheaply after the microprocessor based system was in place. I think trying to implement something like that without the microprocessor would be unbelievably complex, cost prohibitive, and fraught with problems.
Sounds like a solid vote of confidence regarding the use of microprocessors.

If something can give you more control over a system - and give you more options and ability to enhance or "tweak" the system - then IMO that is a significant factor in determining whether it will one day become the norm. And if it happens to cost much less to mass produce than the thing(s) it is replacing, then that makes its eventual implementation pretty much inevitable.
 
  #15  
Old 09-01-13, 05:07 PM
Member
Join Date: Mar 2006
Location: Wet side of Washington state.
Posts: 18,482
Received 25 Votes on 19 Posts
Thank you, Zoesdad, always good to have the information from the people that actually design these things.

Rockledge, I hope I haven't given the impression that I am against the microprocessor controls because I am actually all in favor of them. I saw one job that just had me banging my head against the wall maybe fifteen years ago. The company was doing an upgrade on the boiler controls for the main power plant. They were installing digital readouts for various pressure and temperature controls so needed the electronic transmitters for the readouts. They could have EASILY incorporated digital switches within the readouts but instead chose to use MECHANICAL switches instead. It required a huge amount of redundant wiring in addition to the old technology switches that could be easily bypassed by just about anyone. Some of these were on safety devices so the integrated digital systems were actually safer but for some unknown reason, probably ignorance on the part of the project engineer, they went with the old tech when the modern would have been less expensive and safer. Many thing on that job were just plain bassackwards. I don't think it was ten years and they completely redid it but I have no idea of what the changes were as it was after I retired.

I'd tell you about what I did but alas, much of that is now gone as well.
 
  #16  
Old 09-01-13, 08:30 PM
Member
Join Date: Nov 2011
Location: USA
Posts: 408
Received 1 Vote on 1 Post
Rockledge, I hope I haven't given the impression that I am against the microprocessor controls because I am actually all in favor of them.
That's not the impression I got. It doesn't come across to me that you are against microprocessor controls. It seems to me that you are making the point that a lot of what is being done today with the microprocessor in the name of "energy efficiency" was doable many moons ago with more simple controls.

Similarly, I don't want to be perceived as someone who blindly embraces all the new-fangled HVAC computerized controls and components just because they incorporate some sort of microprocessor.
It all comes down to whether or not they actually work, what options they give me, and how reliable they are. For example, if I have to replace a "board" every other year at $$$$ a pop, then I'm not gonna want anything to do with them.
 
  #17  
Old 09-01-13, 11:15 PM
Member
Join Date: Mar 2006
Location: Wet side of Washington state.
Posts: 18,482
Received 25 Votes on 19 Posts
Fair enough but I wouldn't characterize the controls of the past as simple by any stretch of the imagination. They had their problems just as much as any other control, maybe more since they were far more intolerant of dust or dirt than the electronic controls. Both the pure mechanical and the pneumatic controls (I haven't mentioned them) have one thing that can be both a godsend and cause a person to tear their hair out and that is that they can often work to a degree even when not working properly. Electronic controls almost always work exactly as programmed or not at all; no middle ground of "sort of works" for them. No question that the electronic controls are almost maintenance free when properly applied, something that cannot be said for their predecessors.

I did a lot of work using Powers (manufacturer) model 535 PID (proportional, Integral, Derivative) controllers. These were a basic workhorse control and could be configures for several different input types and up to four outputs of different types. The basic unit cost around $800 and it was worth every penny. These units were almost bulletproof. Shortly after their introduction the Asian manufacturers started building their own PID controllers and offering them for considerably lower prices I remember a Fuji unit that had a more limited number of options in any particular configuration that sold for around $200. The Fuji unit was fairly good but it couldn't hold a candle to the Powers. Now you can buy a basic electronic PID controller off of eBay for around $25 as long as you don't need all the bells and whistles. I will go on record though that the eBay specials WILL crap out long before the Powers unit.

THIS is where I see so many microprocessor controls today. The manufacturer of the finished product, let's say a boiler, buys the basic control for $25. He then adds the input devices, most generally thermistors for temperature measurement and maybe a cheap pressure transmitter as well. All these components are of the cheapest quality that can be found and so their basic accuracy, repeatability and most important, reliability are all pretty low on the scale. But they are usually good enough to replace the mechanical analog gauges and switches so who cares. The manufacturer give the whole unit a one year parts and labor guarantee and maybe a ten year guarantee on the heat exchanger portion. Two years down the road the microprocessor craps out because, well, because it is a poorly made module with low quality parts. It is almost impossible to repair it in the field for many reasons so the whole board is replaced at a cost usually ten times what the boiler manufacturer originally paid for the thing. The foreign manufacturer of the processor sells more crappy units and the boiler manufacturer makes a killing selling replacement parts. The poor homeowner is left with a mediocre product and a whopping repair bill. I would far prefer to have a QUALITY microprocessor that is built of high quality parts and made to last in a rugged environment even if it meant a $100 dollar increase in the price of the $3,000 dollar boiler. The manufacturer could then offer a full ten years parts AND labor warranty knowing that there would be almost no claims. It would be a huge selling point but so many people make the decision on what to buy based on initial price alone it may not end up being practical for the manufacturer.

Okay, enough of my ranting, my hands hurt and I need to cleanout the cat box.
 
Reply

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Thread Tools
Search this Thread
Display Modes
 
Ask a Question
Question Title:
Description:
Your question will be posted in: