Control Board

[Beekster]

I have been passing information onto my brother-in-law to get his
feedback on our 'problem'. He is super-experienced with starters and
with programmable logic controllers. Here's what he has to say about
our control board--mine has the led lights:

"I looked the control wiring schematics. It is a pretty simple system
with only a little bit of I/O. I checked into it, if push came to
shove the main proprietary control board could be replaced with off
the shelf AutomationDirect modular PLC hardware for about $600. The
programming software is $395. The manual pretty much explains how the
various cycles work, and some of the people that worked for Hallowell
could probably fill in the missing pieces. The advantage of course is
that you would have absolute control of how it worked and could tweak
anything you wanted and even monitor everything from a PC, including
logging data. So as long as the somebody is building the compressors
you should be able to keep the system running virtually forever."

Mike

[Tom Berkey]

Interesting idea, but it would require knowledge that is not in any of
the manuals, that is to say the timing of the various logic states
that are necessary to operate the overall system. Certainly, the
communication capability would be a plus in determining the state of
the system and in monitoring the "health" of the overall system.
Unless one had a spare system to play with, it would be a large
undertaking and few of us have the expertise (or funds) to take
something like this on.

Tom

[Kris Heikkila]

I may be able to offer even more on the subject. My neighbor INVENTED
the PLC. ... Andover controls, ModiCon industries, Flavors Inc. ...
All his.
We shall see....
Kris

[Chris]

I've brought this up before. I think most owners think the system's
operation is much more complex than it actually is. There are only
6 outputs and only 3 sensor inputs beside the thermostat and high and
low pressure cut out switches. I work with micro pc's programmed with
cf cards, together with an i/o board and you are under $200. The
programming isn't that difficult but I doubt control boards will
really be a problem. With all the failed units I think there will be
plenty of used boards around as compressor failure seems to be the
last straw for most. Some misunderstand also thinking the controller
could indicate health of the system, all the board knows are the
thermostat stage 1 and 2 on or off and the 3 temp sensors. It isn't
that complex.
Also previously mentioned is there are several other updated versions
of this system all of which use pressure switches in place of temp
sensors. These would provide more information that could be used to
operate the system more optimally.

[Bob Dobbs]

On Jul 19, 4:33 pm, Beekster <beekst...@gmail.com> wrote:

> "I looked the  control wiring schematics. It is a pretty simple system
> with only a little bit of I/O.  I checked into it, if push came to
> shove the main proprietary control board could be replaced with off
> the shelf AutomationDirect modular PLC hardware for about $600. The
> programming software is $395.  The manual pretty much explains how the
> various cycles work, and some of the people that worked for Hallowell
> could probably fill in the missing pieces.  The advantage of course is
> that you would have absolute control of how it worked and could tweak
> anything you wanted and even monitor everything from a PC, including
> logging data. So as long as the somebody is building the compressors
> you should  be able to keep the system running virtually forever."

The original controller boards have a communications port (TTL level
RS232 9600 8E1, and with adapter cards can "speak" standard
RS232, RS485, USB, and ethernet). Among the available data is
are the last 6 state machine decisions, inputs, temperatures,
outputs, state machine data, internal timers, and the ability to
change any numeric parameter in the system (temperature limits,
intervals, etc).

[Tom Berkey]

Thanks Bob - does your comment apply to ALL of the boards produced by your company? Are the adapter cards you mention available from your company, or from some other source? To your knowledge, has anyone attempted to implement this interface and, if so, can you provide a reference?

When I referred to "health" of the system in a previous message, I perhaps should have used the term "status", although I think most readers understood my context. I would like to be able to remotely determine the status of my ACHP and I have attempted to do this indirectly by monitoring the current consumed by the AHU and the outdoor compressor, but this is a relatively crude indicator and less than ideal. The current measurements can tell me how much power is being consumed (which is useful) but do not speak directly to the parameters that are relevant to the operation of the ACHP. Any further detail that you can provide would be useful.

Tom

[David Friedman]

This sounds like valuable info - can we start a new topic on the savemyacadia.org website re Remote monitoring of Acadia's.  Perhaps bob can weigh in with some simple PC based code and a specific com device and write up some info as to how he might see using the system.

 

David

[Tom Berkey]

Following "Bob Dobb's" hint, I spent some time on the AthenaControls website and have found what may be the interface cards that he was referencing in his recent note. Looking at the Athena control board (v1.06) in my ACHP, I note that there is a 10 pin "Comm Header" as well as four holes in the printed circuit card where an external card could be mounted.  According to the 264(!) page manual that is available on the Athena website, "The 2C and 4C controllers are control boards that can perform PID or On/Off temperature control for two zones (2C) or up to four zones (4C)." A quick Google search indicates the PID refers to Proportional, Integral, Derivative and is a term used in the Control community (also explained in the Athena .pdf).  As noted by Bob, the board can speak RS485, RS232, Ethernet or USB protocols and this must be specified at the time of order. Programming the board does not appear to be trivial; I believe that it would take some considerable effort to become fully proficient in learning how to use the board. I note that it does interface with Windows computers. It would be useful to see the board in action ala Gabe's YouTube videos. Finally, I could not find a price for this part on the website, altho it may well be there. 

Here is the relevant link: http://athenacontrols.com/pages/products/loop_pages/4cseries.tpl  To see the gory details, click on the link embedded on this page. 

[Ernst den Broeder]

Bob,

Ok, I'm just plain curious now. (and bored I guess)

COMM. HEADER

PIN 1: RX (of CPU)

PIN 2: TX (of CPU)

PIN 3: some kind of input on PORTC.1 of the AVR - pulled high to 5V through 10k

PIN 4: 5V

PIN 5: GND

PIN 6: NOT USED

PIN 7: NOT USED

PIN 8: NOT USED

PIN 9: NOT USED

PIN10: NOT USED

I tried hooking up my toys but I get nothing out of the port as-is, maybe I need to tie pin 3 to GND to enable it.  Bob, does the serial console on the software 'speak' plain ASCII and provide a CLI or does it expect some kind of binary data to interact with it?

The ISP header (J6) is the standard 6 pin layout for AVR micro controllers (I need to make an adapter for my 10 pin header'd programmer before I can see if the flash inside the AVR has been 'locked').  If it's not locked, we'll have a way to put 1.07 onto other people's boards...

Ernst

[Ernst den Broeder]

I'm probably answering part of my own question here, but after looking around a little bit on Athena's website, it looks like there is a good chance this board is using a protocol called Modbus.  Interesting (to me anyway)!   

[Tom Berkey]

And you would be correct! What can you tell us (me?) about the MODBUS protocol??

Tom

[Beekster]

More from my brother-in-law:

"I suggested an off the shelf PLC rather than a micro PC route because
my experience with micro PCs as control devices has been bad. I
actually built 3 micro PCs using Advantech boards, a 1500 nit sunlight
viewable 480 x 640 display with a capacitive touch screen enclosed in
a Nema 4X fiberglass housing with a peltier effect thermoelectric
cooler to allow operation in direct sunlight at temperatures up to
around 130F. We had a touchscreen driver card fail, and the new
replacement wouldn’t work without a BIOS upgrade, after one year! The
next install of the software on the second one wouldn’t work. Hours
and hours of troubleshooting finally determined that one of 600 files
required for the Windows XP minimal installation was corrupted on the
Microsoft CDs! PCs change too fast, you are constantly chasing the
changes to keep things working. On the other hand the PLCs I replaced
those micro PCs with have run with little problem for about 10 years
now. The programming software is simple ladder logic, and I can talk
the customer through programming changes over the phone. I have PLCs
out there running for the past 18 years with no problems. Proprietary
boards like in your unit are also usually very robust, because they
can be optimized for the purpose and they seldom change much. The
gotcha is the proprietary part if the manufacturer goes belly up. We
all know that never happens!

As you know I build wastewater treatment systems using reverse osmosis
and ultrafiltration. At last count the system I work on has about 168
inputs, including 3 temperature transmitters, 8 pressure transmitters,
2 dissolved oxygen, 2 pH, 2 Orp, 4 Level transmitters and about 30
limit switches. This controls 12 pumps 3 blowers, and 16 valves. In
the industrial world this is a VERY small system! So doing controls
for a single HVAC unit is not all that difficult.

Tom mentions in his post about monitoring amps on his compressor and
he is right, amps is a very crude measurement, especially on single
phase power. You really need to measure watts to monitor what the
compressor load is. Watts are a function of amps, volts and power
factor simultaneously. Amps don’t actually indicate load. This also
brings up data logging. My experience with data logging has always
been an eye opener, because I always find out something that was
hidden from the casual observation of data. Recently a data log on
some simple pumps that moved water from tank to tank showed that one
pump was actually worn out, and the pumps often short cycled because
the operation of the pumps was just plain wrong. It had only been
that way for three years! If you really want to know what one of
these systems is doing you really need to do data logging so you can
see exactly whether the dog is wagging the tail, or the tail is
wagging the dog! You can use the PLC programming software to data log
too.

That Com port on the factory boards is interesting. Does anyone
actually have the interface card, know how the data stream is
formatted, and is there PC software that gives read/write access, and
logging? It certainly would go a long ways towards understanding
what kills the compressors.

Chris mentions pressure switches, I think he meant analog pressure
transmitters, not switches. Again that would be a very good idea to
understand the compressor problem. Temperature changes are very slow,
while pressure changes are nearly instantaneous. So a pressure sensor
could indicate a problem almost instantly and respond to it before the
compressor was damaged. Probably overkill for every unit, but if you
are trying to understand the problems it would be extremely useful."

[Bob Dobbs]

On Jul 21, 8:29 pm, Ernst den Broeder <erns...@gmail.com> wrote:
> Bob,
>
> Ok, I'm just plain curious now. (and bored I guess)
>
> COMM. HEADER
> PIN 1: RX (of CPU)
> PIN 2: TX (of CPU)
> PIN 3: some kind of input on PORTC.1 of the AVR - pulled high to 5V through 10k
> PIN 4: 5V
> PIN 5: GND
> PIN 6: NOT USED
> PIN 7: NOT USED
> PIN 8: NOT USED
> PIN 9: NOT USED
> PIN10: NOT USED

Pin 3 is an output -- RS485 xmit enable.

> The ISP header (J6) is the standard 6 pin layout for AVR micro controllers (I need to make an adapter for my 10 pin
> header'd programmer before I can see if the flash inside the AVR has been 'locked').  If it's not locked, we'll have a
> way to put 1.07 onto other people's boards...

It is locked. You can erase, but not read.

[Bob Dobbs]

On Jul 22, 1:13 am, Tom Berkey <tom.ber...@gmail.com> wrote:
> And you would be correct! What can you tell us (me?) about the MODBUS
> protocol??

http://www.modbus-ida.org/

I'll see if I can get permission to list some of the registers. You
*REALLY* do not want to
play with them all, you can blow your compressor into shrapnel that
will take out anything
close to it.

[one mad dealer]

Bob,

We have a circuit board discussion going on the new savemyacadia.org
website. We would appreciate it if you would register and weigh in on
those discussions. Thanks.

Gabe

On Jul 22, 11:35 am, Bob Dobbs <controlsystemprogram...@gmail.com>
wrote:

[Ernst den Broeder]

Agreed, I don't want to modify any of the values, but being able to read specific registers for the purpose of data logging and temperature sensor diagnosis would be most awesome.

[David Friedman]

Ernst,

 

If you follow through on a complete design for a data logging system allowing for remote monitoring - it would be great to fully document it and offer the information on the savemyacadia,org

 

I have been following the discussion but not in detail and it seems as if you are  homing in on an awsome solution that would allow for a user to get data on his performance and health of his system from a remote PC at his alternate home.

 

Along the line of "concern for remote operation" there is a flaw in the current Acadia setup that Gabe and I have worked out a solution for.  If the current system loses power due to a short on the outside unit - the TS can not deliver strip heat to save the house from freezing.  This is fixable by a rewiring of the 24 volt power supplied and we will offer a FWO do due this.  Look for it on savemyacadia.org.

 

If you get to a usable remote monitoring configuration of hardware and software - please post it to the new site.

 

David

[Tom Berkey]

Perhaps "Bob" give us an idea of what costs to anyone interested to go down this path would be incurred & if we are looking at the correct product. I have previously invested in a Brultech product that allows current measurement but as discussed previously, this is both an indirect measurement and not particularly enlightening, particularly if a fault of some kind occurs. It does allow the user to see the time history of the current being used (in my case) by the electric heating elements and the outdoor compressor (I have only 2 split 60A sensors). 

Any further detail and or references regarding remote monitoring products/applications would certainly be appreciated.

Tom 

[David Friedman]

Tom

Another almost funny example of value of knowing what actual state system is in occurred yesterday.  The power company disconnected my power based on my Heat Smart rate for four hours so the house got hot and it was a good chance to really observe the system as it recooled.

 

I found to my amazement and horror that when my TS for the main zone called for one degree of heat and the Y1 signal was active at the outside unit - the primary two relay was fired.  It turned out that th ewiring from my PC to the contactors was reversed.  The system still sort of worke d- but in the winter I must have never gotten the full benefit since at cold temps I ended up riring primary one along with tghe booster - rather than the correct combination of primary two and booster.

 

A private conversation with Michael Barrick indicated that he has found many units (Perhaps with the new PC boards) wired incorrectly so I think I will start a new topic on the savemyacadia.org site covering that curious and subtle factory mistake.

 

David

[Chris Papademetrious]

David,

Great hook to get folks over to the forum. :)

- Chris

[Chris]

My contactors are wired opposite the control board schematic too. The
owners manual shows yellow connected to 1cyl and blue to 2cyl on the
acadia board. My blue wire is longer than the yellow and yellow wont
quite reach contactor 1. The wires on my controller were switched
from the factory. Blue is connected to 1cyl on the board and yellow
is connected to 2cyl. These are the wires that connect to the side of
the contactors. Mine has blue on the left and black on the right of
the pri1 contactor and yellow on the left and black on the right of
the pri2 contactor. Certainly confusing. The new siemans contactor
makes it even worse as you cannot see the contact physically engaging
because it is covered.

[David Friedman]

This condition causes the wrong operation calling for cylinder two first and then cylinder one.  Worse yet - in heating mode - the system never will develop the max compression to get the max since that requires both cylider two and booster.  I carefully had system call for stage one heat on my main TS and observed that Y1 was high at the outdoor unit - and then measured voltages across the primary one and two coils verifying that the wrong cil (primary two) was being energized.  As far as I can tell it was a factory screwup.

 

 

This problem didn't exist with the older wiring harness associated with the older board which had single wires as opposed to pairs going from PC to contactors.

 

David

[Chris]

My test board came with it's original harnesses. The board is v1.07.
Here, the 1cyl output is yellow and 2cyl is blue as it should be. The
yellow wire is longer than the blue to reach the pri1 contactor. Both
my original and my test boards were wired correctly, although my
original used the wrong wire colors. It is easy to tell if this is
right with an ammeter. If 1cyl is lit on your board your current
should be around 6 amp. 2cyl draws around 12. I apologize for my
slowness, but I continue to be confused by your circuit references, I
wonder if I am the only one. Please explain what you mean by y1 is
high? Was it smoking something? Do you mean there was voltage
present here? Also I'm afraid I don't know how to make my unit call
for stage 1 carefully.
The moral is don't depend on wire colors in schematics from hallowell.

[David Friedman]

Chris

 

I too feel a bit of confusion re wiring and will try to be  as clear as possible when talking about wiring and contactor states.  Perhaps it might do us all a bit of good when working on our beloved Acadia's to be a bit high.

 

When I said Y1 high I meant it measures 24 v ac.  I measured this voltage at the connector strip below the PC board and also at the primary contactor.  When I said I carefully called for stage one cooling - I meant I nudged the main TS (second zone TS was set to a non operating temp) so as to call for just one degree of cooling so as to have the TS emit a Y1 signal. Of course I could have just jumpered the 24 volts as needed at the outdoor unit - but I wanted to see the TS create the correct state at the outdoor unit.

 

I happened to be debugging a second "confusion" in my HVAC ssetup - namely that the second zone TS supplied (An American standard with a number ending in aa) turned out to not be a two zone TS (programming screen doesn't allow multi stage) and when the second zone called for AC that TS emmitted both a Y1 and a Y2 totally confusing me since the alpha numeric screen kept ending up in c2 as the cooling state despite my only calling for stage one at the TS.

 

After realizing that I had an inferior TS, and after discussion with Gabe discovering that HVAC practice often considers only using a single stage TS for a smaller zone - I removed the Y2 (blue wire) connection from my second zone TS down at the zone control so that only my Hallowell primary TS can call for second stage.

 

However when I (as mentioned above) made sure that the call for AC came from the main zone with the Hallowell TS and only emitted stage 1 (Y1 out from ts) - I found that the PRI-2 contactor coil had 24 volts powering the heavy Yellow Bristol wire to the R terminal of the compressor.  Assuming that Bristol hasn't swapped wire colors at the compressor - I concluded that the two - two wire pair outputs from the single connector on the Invensys PC were swapped as they ran to the contactor coils. The colors on the wire pirs matched the physical connector positions as the wires left the PC at the connector which feeds all of the relay outputs from the PC.

 

After a good nights rest - with a clear head - I repeated the test of invoking minimal AC call which  produced only Y1 and assured myself that the only the PR-1 contactor has 24 volts applied, and that this puts power on the heavy blue wire going directly to the compressor from the PR-1 contactor.

 

I don't have a clip on ampmeter so I couldn't check current - but voltage from R to L2 was about 220 which would have been  adequate to operate the old PR-1 potential relay (I have a USMR setup which is voltage independent) and indicates that I am truly running PR-1.

 

Chris - given that there was a wire color error with the harness for early "blink" boards, perhaps this explains why the manufacturing error of connecting the correctly colored and correctly wired at the PC connector ended up allowing Hallowell mnfg. to swap the cable pairs between the two contactors.

 

David Friedman

[David Friedman]

In re-reading the post - please note that I had fixed the wiring set up last night - and then not quite trusting myself - went out this morning to verify that the now correct wiring produces the correct contactor/compressor state.

This plus the removal of the blue Y2 wire from the single stage second zone TS has m Acadia producing the correct compression corresonding to TS operation as desired.

 

As a side not.  I checked the air handler jumpers - and found that my HVAC tech had not set these correctly and also changed these.  The level of air flow now seems to correspond correctly to the TS call.

On Sat, Jul 23, 2011 at 11:00 PM, Chris <cpete...@yahoo.com> wrote:

[Brian Goates]

David or Gabe, can we get a descriptive once over on exactly what needs to be checked/changed with this wiring conundrum and the air handling jumpers? My AH output is always the same so I have something off.

 

Is the AH jumper setting at the AH or at the OD?

 

Also I have completed the UMSR upgrade and following the wring in those instructions , so does this new revelation counter those instructions?

 

It may have been mentioned but there has been quite the discussion on these two items!

 

 

Tks, Brian

[Tom Berkey]

I believe that much of the confusion that arises in these discussions could be addressed by everyone being more precise in what they are talking about. Case in point is David's constant reference to the PC - I assume he is using this as a shorthand notation to the system control board. However, someone just joining these discussion might think he's talking about a printed circuit board (which it is) or a personal computer (which it most certainly is NOT). Also, some tend to talk about "ohming" something out (yeah, I also saw this used in a Hallowell manual), but the noun ohm is NOT a verb and shouldn't be used as such. More recently, a voltage was referred to as being "high" which led to some confusion. Low and high states are commonly used in digital electronics, whereas the HVAC systems are normally 24 VAC. Mind you, we have a digital control board, so there are digital systems at play here. Finally, Gmail has a spell-checker facility which I would recommend that we all use. 

End of rant.

Tom

[David Friedman]

We are helping out our daughter and son-in law this weekend in Boston - but I will post this as a topic on savemyacadia.org on monday.  The USMR instructions were written so as to refer to physical locations and pin numbers so there is no confusion.  The error is quite non destructive in AC mode so getting it right is not critical - but now that I have straightened out my unit - I feel that its performance is improved.
David

[Chris]

"getting it right is not critical"

Huh?? How do you know what you are doing if you don't have basic
tools?
You're telling us how to fix ours?? This is just your guess work?
Good luck everyone!

[Bob Dobbs]

On Jul 22, 11:35 am, Bob Dobbs <controlsystemprogram...@gmail.com>
wrote:

I've spoken to management, and because of lawsuit with PH,
register list and boards will not be available at this time.

[David Friedman]

Mr. Peters,

 

As usual your comment is misplaced,  wrong,   and terribly rude.  You are clearly a disturbed and angry person who has repedtedly violated the behavior rules that should govern this site.  You should have been cut off long ago for rude and disrespectful remarks.

 

If the PR-! and PR-2 are switched as were  mine  until I rewired - the consequences are rather mild and won't destroy the compressor.  Of course we all want the system to operate as designed - and with PR-2 wired to run  instead of PR-1-- the  AC has excessive initial cold air flow but it works. 

 

My comment was intended as an excuse for my not getting the full write up done as soon as I wanted to. I am doing that today - and will post it as a forum topic on savemyacadia.org.

 

As to requirement for an ampmeter to determine  how the system is wired -  for me at least - my voltmeter tells me all I need to know and I can tell whether I am running cyl-1 or cyl-2 based on the coil voltages.

[Ernst den Broeder]

Hi David,

Since BobDobbs has reported that Athena Controls cannot (at this time anyway) release appropriate documentation to us that would aid in making a way to datalog the Athena control board, I will not be spending any further effort on this idea.

For now, it would be much easier to use an off-the-shelf data logger and graft it in parallel to the Athena control board.

regards,
Ernst