Weil Mclain Ser. 3 manual reset lockouts on F1/2 & O1/2 temp. diff.

Reply

  #1  
Old 01-15-15, 07:46 AM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
Weil Mclain Ser. 3 manual reset lockouts on F1/2 & O1/2 temp. diff.

Hi All,

Curious if anyone knows a likely cause for a recent flurry of manual reset lockouts our Series 3 natural gas Ultra 105 has experienced. This began at the end of last year (prior to the recent, ridiculous cold we experienced here in CT). The boiler was last serviced in May (std. maintenance - all nominal), has run fine since installation in 2009, is driving three zones of baseboard heat, target temp. set at 180º F, no DHW (yet), no outside temp sensor installed, version 1.0 control software.

Late last year the system threw two manual reset lockouts in a row on "Temperature Sensor". The system ran fine for a few weeks and then on the 6th began misbehaving again. I recorded the temps for each of these later instances and noticed two things:

1. In each case the temp diff between the FLUE1 / FLUE2 AND the OUT1 / OUT2 were considerably greater than the 10º mentioned in the manual - ranging from a 15 to 20º difference. With each lockout, both of these dual sensors are out of spec., that is, I never see a F1/F2 diff without seeing a corresponding O1/O2 diff. These diffs have all occurred in the 105-156º range. I can post the actual numbers if this is useful.

2. This behavior only occurs when coming up from ambient temp. We normally keep our living area around 67-69º (cooler at night) and the house is well-insulated, so the boiler doesn't actually run all that much even when it's cold out, and is off for considerable stretches. During these off cycles, the internal temps all drop to the 60's and the lockout only occurs when bringing the system up from that level. Of course, this is usually at night when we're asleep, so a lockout means waking up to a cold house.

I have verified #2 by tweaking our main floor thermostat up 2º, which - at least with the recent cold - has shortened the down time on the cycle, preventing internal temps from dropping much below about 130º before a zone calls for heat. The system has run without a lockout since. In fact, if we normally kept our house a little warmer, we probably wouldn't have noticed this issue until spring. Also, when I visually monitor the F1/F2 and O1/O2 temps during operation, they always look fine (diff. within 1-2º).

My contractor took a look, measured some resistances and didn't find anything obvious (it's pretty intermittent, after all). He talked to WM's phone tech support, who was absolutely no help. The WM tech chose to obsess over the placement of the external, pipe-mounted supply/return sensors (???), which have been just fine for 5 years, and ignored the F1/F2 O1/O2 temp. diffs, reported with the errors, entirely. We decided not to futz with it for the moment, since the system's providing heat reliably.

I've built and programmed PID-controlled IR heat reference systems in the past, so I have a little more more than a passing familiarity with the electronics and algorithms involved here. The key aspect, to me, is that when this problem occurs, both of the dual sensors (F1/F2 and O1/O2) are reported as out-of-spec. Unless a failure of one sensor is actually affecting the other electrically, I have a hard time believing that both of these dual sensors decided to fail, intermittently, in exactly the same way at exactly the same times, so I suspect whatever is supplying their bias voltage. The obvious first step (to me) would be to replace both dual sensors and see if the problem still occurred, which I suspect that it will. I'm not familiar enough with the design of this system to know what the next step would be. I'm hoping someone here may have seen something like this before, in which case we may be able to cut to the chase without the intermediate steps (and expense).

Any and all suggestions greatly appreciated. TIA.

cheers,

Ron
 
Sponsored Links
  #2  
Old 01-18-15, 06:54 AM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
*bump*

Hmmm... so... no thoughts on why the FLUE1/FLUE2 and OUT1/OUT2 sensors would suddenly start going out-of-spec intermittently? Maybe just time for a new control board.

cheers,

Ron
 
  #3  
Old 01-18-15, 08:53 AM
Grady's Avatar
Forum Topic Moderator
Join Date: Jan 2005
Location: Delaware, The First State
Posts: 13,936
Received 1 Vote on 1 Post
Having more experience with electonics than we do, you are probably the more qualified to try troubleshooting that part of the system.
 
  #4  
Old 01-18-15, 09:20 AM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
Thanks Grady. I understand the system, in principle, but I'll leave the actual troubleshooting to the experienced (and bonded) pros.

Since my contractor is stumped, I was just hoping that perhaps someone out there had seen symptoms similar to this on a boiler in this series, and might have suggestions for an approach.

In other news, our cats are enjoying the slightly warmer temps. ;-)

thanks again,

Ron
 
  #5  
Old 01-18-15, 09:30 AM
Grady's Avatar
Forum Topic Moderator
Join Date: Jan 2005
Location: Delaware, The First State
Posts: 13,936
Received 1 Vote on 1 Post
Ron, one thing that just came to what little mind I have left is to make sure ALL ground connections are clean & tight. We often see ground problems cause crazy things to happen with gas furnaces which almost all have circuit boards & various sensors.
Also check voltages: Hot-neutral; hot-ground; neutral-ground. If you find any voltage over ~ 1 volt between neutal & ground, that could be the source of your trouble. Has anyone done any electrical work in the house recently?
 
  #6  
Old 01-18-15, 10:18 AM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
Thanks Grady. That, I can check.

No electrical work done for well over a year now

cheers,

Ron
 
  #7  
Old 01-18-15, 11:13 AM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 2 Votes on 2 Posts
Hi Ron,

I have a hard time believing that both of these dual sensors decided to fail, intermittently, in exactly the same way at exactly the same times
I would first be investigating what ELSE might be wrong.

Perhaps there's some other issue going on that the sensors are correctly reporting a fault.

In other words, you seem to be stuck on blaming the messenger.

Make sure there's not a root cause reason that they are reporting!
 
  #8  
Old 01-18-15, 12:20 PM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
Thanks Trooper.

What else might be wrong? Can you think of specific examples I can check or have my contractor look at?

Is there actually some known condition, some root cause reason in a boiler of this type that will result in two different, redundant sensors - in completely different parts of the system (one measuring air (FLUE1/2), the other measuring water/manifold temp(OUT1/2)) - to both begin consistently, simultaneously reading "unbalanced" in this manner, at a fairly consistent point in the heating cycle, and tripping out the boiler?

This is what I'm hoping to learn, since neither my contractor, nor the install/maintenance manual nor the phone support at WM have been any help at all.

The manual is a problem right at the outset. It describes this condition - i.e., "...sensor not balanced" - as one that should cause an automatic reset condition (p. 85 in my series 3 book). The "Reset action" described is "Reset if sensors balance". Problem is, the boiler doesn't automatically reset under these conditions; instead it goes into manual lockout, despite the fact that this condition isn't listed as one of those that should result in a manual reset lockout.

The stored temps for that error, when it occurs, clearly show both sets of sensors "not balanced" (i.e., >10º discrepancy), but when I check the 'live' temperatures immediately following one of these manual lockout events, both pairs of sensors (FLUE1/FLUE2 & OUT1/OUT2) read perfectly balanced. And while monitoring these temps during a heating cycle, I never see them get more than 1º apart.

Meanwhile, the only "Temperature Sensor" condition that should put the boiler into manual reset lockout (according to the list on p.87 in my book) is "Temperature sensors Hx inlet or outlet, or flue sensor is open or shorted ". Unless the stored temp. values are a lie, then neither of these conditions can be the case.

So I'm not stuck on anything, exactly. I'm just looking for an approach to resolving this based on the evidence available. While anything's possible, it's not my experience that two redundant sensors in different parts of a system like this will suddenly and consistently start failing in exactly the same way at exactly the same times. Much more likely that the problem is in something they have in common. That's what I'm wondering. But without a schematic of the control board, there's no way to know what that might be. According to the manual, they both plug into the same socket on the control board (P8). That's all the detail that's available.

Anyway, if no one has an alternate suggestion, then I'll probably just go ahead and turn the house heat back down to our normal setting. See if the problem resurfaces. If it does, I'll just arbitrarily have my contractor order a new set of sensors, swap them out and go from there, based on what happens next.

thanks!

Ron
 
  #9  
Old 01-18-15, 02:43 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 2 Votes on 2 Posts
I'm reviewing the manual and cogitating right now...

One of the things I always try, which works some of the time, is plugging and unplugging associated connectors. I'm sure you have probably already done this.

You're saying that the difference that the control is puking on is between each of the dual sensor pairs, not a difference between the two pairs of sensors... right? Just want to be clear what I should be thinking about...
 
  #10  
Old 01-18-15, 03:11 PM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
Thanks Trooper.

Yes, these two probes actually each have two sensors (three wires coming off of each, IIRC).

The flue probe has FLUE1 and FLUE2 and the hx Out probe has OUT1 and OUT2.

According to the manual (under the auto reset conditions on p. 85), if the temp. difference between FLUE1 & FLUE2 becomes unbalanced by > 10º, an auto reset will occur. Same with OUT1 & OUT2.

The system's behavior doesn't match the documentation, however, because the unbalance condition is causing a manual reset lockout, not an auto reset condition.

When this behavior occurs, BOTH of these sensor pairs are out of spec - at least as recorded in the data for the error. Again, if it's useful, I can post all the numbers recorded. It's never just one pair or the other. Thus, my thought that the issue is related to whatever is providing their bias voltage.

thanks again,

Ron
 
  #11  
Old 01-18-15, 03:45 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 2 Votes on 2 Posts
Thus, my thought that the issue is related to whatever is providing their bias voltage.
Yes, or a common ground issue.

They both plug into the control board right next to each other. It's very possible that even something such as a bad solder joint could be at fault...

I wonder if the sensor inputs are 'multiplexed' into the microprocessor, or if they are discrete inputs that are 'polled' sequentially...

Still cogitating...
 
  #12  
Old 01-18-15, 04:20 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 2 Votes on 2 Posts
OK, so what have the two sensors got in common that could cause them to both fritz at the same time?

They're both on the same connector:



Here's a 'what if' to think about:

What if there's 'crud' (technical term for fly poop and flooby dust) bridging pins 3 and 4 of that connector?

That would cause one each of the pair to read whack, wouldn't it?
 
  #13  
Old 01-18-15, 04:25 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 2 Votes on 2 Posts
an auto reset will occur.
I wonder though, how many 'auto resets' will be attempted before it gives up in frustration?

Even microprocessors can get frustrated! (never anthropomorphize inanimate objects, they don't like it when you do that!)

So say it sees a bad reading and auto resets... and still the reading is bad... and again...

If the designers saw fit to include redundant sensors, surely they would have designed a watchdog into the firmware.
 
  #14  
Old 01-18-15, 04:27 PM
Member
Join Date: Dec 2014
Location: United States
Posts: 17
Received 0 Votes on 0 Posts
pull the condensate trap and blow out what was once part of your boiler.
 
  #15  
Old 01-18-15, 04:29 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 2 Votes on 2 Posts
pull the condensate trap and blow out what was once part of your boiler.
Completely irrelevant to the discussion at hand.
 
  #16  
Old 01-18-15, 04:42 PM
Member
Join Date: Dec 2014
Location: United States
Posts: 17
Received 0 Votes on 0 Posts
it's a condensing boiler. blocked trap drives temp sensors crazy. and if it's Lp it will fail!
 
  #17  
Old 01-18-15, 04:47 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 2 Votes on 2 Posts
it's a condensing boiler
Ya don't say!

blocked trap drives temp sensors crazy
I can't disagree, however, if you read the posts you will see that the symptoms clearly do not indicate that this has anything to do with this problem.

But in the interest of completeness, Ron, you have had the boiler properly serviced all these years, is that correct? And your guy checked all this stuff out when he was there?
 
  #18  
Old 01-18-15, 04:53 PM
Member
Join Date: Dec 2014
Location: United States
Posts: 17
Received 0 Votes on 0 Posts
here is one of 20-30 i had to change out in a 4 year period.Name:  ultra.jpg
Views: 1918
Size:  22.9 KB
 
  #19  
Old 01-18-15, 04:59 PM
Member
Join Date: Dec 2014
Location: United States
Posts: 17
Received 0 Votes on 0 Posts
just stuck the tiny screwdriver into the hole where the condensate leaked out because it ate through the cast aluminum heat exchanger. typical for these and others that are cast aluminum.
 
  #20  
Old 01-18-15, 05:04 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 2 Votes on 2 Posts
Yes, we understand that.

Please stick to the issue that Ron is posting about.
 
  #21  
Old 01-18-15, 05:46 PM
Member
Join Date: Dec 2014
Location: United States
Posts: 17
Received 0 Votes on 0 Posts
so you drink the weil-mclain juice then!
 
  #22  
Old 01-18-15, 06:16 PM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
That would cause one each of the pair to read whack, wouldn't it?
Very likely. Seems like the Flooby Effect from this would be pretty consistent, however - at least observable at SOME point, no? I've monitored 'live' temps at length and never seen more than a 1º discrepancy in either sensor pair. The only evidence of this 'unbalance' ever recorded is in the error data for the error instance in question (i.e., stored for recall via front panel prior to manually resetting the unit), and always within a fairly narrow temperature range when the system is coming up to temp from 'cold'. This hasn't happened at all since I goosed our main zone thermostat up by 2º, which prevents the boiler from cooling all the way back down before some zone calls for heat.

On the condensate trap question: the boiler has been religiously serviced. Last scheduled service was in May. Was checked again recently during the service call. Contractor is coming out this week to generate a quote for an indirect DHW system; I'll ask him to check this again, just to be sure.

On the "how many auto resets?" question: the system stores the three most recent error events (of all types - both auto reset and manual lockout). Whenever I respond to a manual lockout, I always check the conditions for any preceding errors and write down all the stored temps. In each of these manual lockout cases, the only error conditions stored prior to the manual lockout error were auto reset events caused by "Temp rise too fast", and the associated auto reset action was successful in all those cases. In terms of timing, there were hours of separation between these stored auto reset events and the subsequent manual lockout events.

Thanks to all!

Ron
 
  #23  
Old 01-18-15, 06:30 PM
Member
Join Date: Dec 2014
Location: United States
Posts: 17
Received 0 Votes on 0 Posts
temp rise too fast is lack of water flow. how is the primary piping? straight up to tees?
 
  #24  
Old 01-18-15, 06:45 PM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
Probably easier to show you...

Supply is on the far right up to the tee; return is on the left, flowing down through the black TACO circ pump.

Name:  BoilerPlumbing.jpg
Views: 1856
Size:  38.8 KB

thanks again,

Ron
 
  #25  
Old 01-18-15, 06:47 PM
Member
Join Date: Dec 2014
Location: United States
Posts: 17
Received 0 Votes on 0 Posts
from that pic that is not piped correctly
 
  #26  
Old 01-18-15, 06:51 PM
Member
Join Date: Dec 2014
Location: United States
Posts: 17
Received 0 Votes on 0 Posts
and that pic only proves that.
 
  #27  
Old 01-18-15, 07:01 PM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
I'm looking at the piping schematic in the manual. Looks correct to me, but I'm not a boiler guy. So hopefully you can describe what's incorrect about it, so I can point that out to my guy when he comes by this week (he didn't plumb it, FWIW). If something needs fixing, we can consider adding that to the DHW system quote.

Name:  PipingSchematic.png
Views: 3060
Size:  30.8 KB

thanks!

Ron
 
  #28  
Old 01-18-15, 09:09 PM
Member
Join Date: Jan 2008
Location: Southeastern Pennsylvania
Posts: 2,947
Received 0 Votes on 0 Posts
hi guys –

Let me put in my two cents,lol. I would not be too confident that the control is designed to actually enter the “auto reset” state verses the “lockout” state for various legitimate error conditions. In my experience many times error condition handling and supporting documentation are given short shrift due to development schedule pressure. In other words I’m saying it is possible that the manual is just plain wrong and that in fact the “lockout” state is entered for sensor(s) imbalance and not the “auto reset” state. But obviously I don’t know. I’m just saying IMHO that could be the case.

Also, page 85 in the manual talks about a “resistance decade box” connected in place of a sensor to double check that the control module is checking the sensor correctly. Could you do that?

If I understand you are saying that (|F1-F2|>10) AND (|O1-O2|>10) at the same point in time which is the point in time of failure. And how could it be that the “F” dual sensor and the “O” dual sensor both fail at the same time. I see what you mean, that sure does seem extremely unlikely. But my only question would be could the failure actually be another temperature sensor and then the processor enters a state in which the F/O sensors information from that point on is not updated in the normal way. In other words, when operating in the error state the “F” dual sensors recordings drift apart say for example as only one of the pair is being updated. A real long shot, lol. But I’ve seen some really weird stuff in my day, lol.

Just seems to me that the error message “Temperature Sensor” does not explicitly point to F/O sensors, but I guess one would have to certainly infer that it meant the F/O sensors if the data clearly showed their deltas out of spec. But it seems to me if the failure is actually another sensor (or something else) the F and O deltas may be a red herring.
 
  #29  
Old 01-18-15, 09:47 PM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
Thanks, zoesdad.

As a software developer, I've spent some time thinking along the same lines as what you outline here - commonly referred to in the vernacular as a ... bug, also sometimes referred to as "really weird stuff". ;-)

Problem is that once I start imagining what sort of bugs may be lurking in the control software for this thing, well, the sky's pretty much the limit, isn't it. If the QA on this model is really that bad (and I've read a good deal from folks who appear to know what they're talking about, who believe this model is way overengineered and undersupported), then I should replace it. I'm just not ready to draw that conclusion yet. That day may come but... well... you know, that's not this day.

Your comments on the documentation, IMHO, are spot on: i.e., it absolutely does not reflect the behavior I've seen. I don't have a dec box (anymore). I'll ask my guy this week about that as well. My sense is that this is not going to help identify an intermittent problem. The sensors read perfectly fine 99.99999% of the time. Replacing with a dec box to verify that the temps match the resistance values won't help catch the 0.000001% that's apparently causing the thing to lock out.

thanks again!

Ron
 
  #30  
Old 01-19-15, 05:27 PM
Member
Join Date: Jan 2008
Location: Southeastern Pennsylvania
Posts: 2,947
Received 0 Votes on 0 Posts
hi Ron –

Gotcha. (I was a developer myself for well over 32 years, mostly fault-tolerant systems for the military/govt.)

I thought maybe the lack of automatic reset afterwards pointed to a software problem, but I guess if it is something in the HW common to both the F/O dual sensors creating a problem as you suspect, I guess a manual reset could and might be the only thing that could clear that, the auto reset continuing to fail until the HW is reset.

But I wonder if the problem could be with a Supply/Return sensor and then all bets are off with the diffs, explaining why the tech acted like this:

The WM tech chose to obsess over the placement of the external, pipe-mounted supply/return sensors (???), which have been just fine for 5 years, and ignored the F1/F2 O1/O2 temp. diffs, reported with the errors, entirely.
In other words, could it be maybe the tech knows a dirty little secret that the diffs are not reliable when a supply/return sensor (albeit ostensibly unrelated) causes a problem. His attitude as you state above is what got me suspicious about the reliability of the diff’s in the first place, but I guess he could be the kind of guy that ignores a lot of things (look forever for your lost keys under the lamppost, because that’s where the light is, lol)

I wonder if it would be possible to force a supply or return sensor problem and see how the system handles that? And maybe even disconnect the F or O sensor and observe what happens. Maybe there would be something surprising and illuminating in the system actions and the display? However, that may be a dangerous thing to do, but I’m sure the opinion of the knowledgeable guys here could be taken as golden advice. Easy for me to say let’s experiment with your system lol. (I can just hear you and Trooper say in unison “You are nuts zoesdad”, lol)

Good luck, I know what you mean, how do you set a trap for that extremely intermittent failure without access to the code.

(We had a bug once in one of our commercial OS’s we modified to make it fault-tolerant for a special project, in which a task sometimes was being restarted with the wrong set of registers, the register values that did not belong to the restarted task but to another task(oh my!). I never did figure out how to trap that one without the trap becoming a major project! Don’t remember what happened.)
 

Last edited by zoesdad; 01-19-15 at 05:48 PM.
  #31  
Old 01-19-15, 07:05 PM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
...the auto reset continuing to fail until the HW is reset.
Well, that might bear consideration if there were any evidence that an auto reset was even attempted (much less failed), but it wasn't - at least no auto reset for "Temperature Sensor" was recorded in the (short) list of errors stored in memory. Furthermore, any auto reset events recorded are typically - no... always - hours prior to the manual reset lockout on F1/F2/O1/O2 unbalance event. As such, they are pretty much, by definition, unrelated to the lockout.

If there were any timing data showing auto reset attempts immediately prior to the manual reset lockout, I'd agree that your scenario bears scrutiny, and I'd be on the phone with W-M asking for supervisors' supervisors until I could get someone to explain why the boiler operation is absolutely nothing like what's documented in the install/maint. manual. But so far I haven't seen any recorded data showing this.

thanks!!

Ron
 
  #32  
Old 01-20-15, 10:06 AM
Member
Join Date: Jan 2008
Location: Southeastern Pennsylvania
Posts: 2,947
Received 0 Votes on 0 Posts
I happened to find this manual not realizing it was apparently an older version. On page 57 it describes in the inset the use of what I think is the dual “O” sensor to be used also for low water cutoff protection. It then says a “manual reset” is in fact required if the diff exceeds 10. But there is a newer manual that does not say anything like that in the same section “Field Wiring”.

Could it be you have the older version of the control and if the “O” diff exceeds 10 a "manual reset" is in fact required? Just a thought. I’m not that confident that I know what they are really talking about and might be getting my sensors mixed up. lol


Old manual
:

http://www.weil-mclain.com/en/multim...ler-manual.pdf

New manual
:

http://www.weil-mclain.com/en/assets/pdf/manual.pdf
 
  #33  
Old 01-20-15, 02:58 PM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
Interesting. Thanks for this research!

The version you refer to as "Old" is identical to the hardcopy version that came with my unit and the PDF I've been using for reference. I searched that PDF for "OUT1", "FLUE1", or any combination of text that would catch a reference to these sensors outside of the tables where they're mentioned. Guess I should have searched for "dual sensor", and I would at least have seen that inset, FWIW. The description in the inset precisely explains what I'm seeing - at least with respect to the data I've recorded. It still doesn't explain what's causing it. We have a separate, external unit for low water sensing/protection (CT code, I believe) that signals the boiler to shut down and trips a rather ear-splitting alarm if there's a problem. It's just below the main circulator (just out-of-sight in the pic posted previously). That thing is tested with each maintenance visit, and has never so much as burped otherwise. I'm going to mention it to my guy again this week, however, just for thoroughness' sake.

The newer manual mentions the same thing the older one does, but doesn't specifically mention how it does this - see the Notices on p. 13 and 21. These notes don't mention anything about which sort of lockout is caused, however. That's actually described in the table on p. 88. Here, too, it claims that an auto reset will take place if the dual probes become re-balanced.


One thing all this does show is that the (older) manual itself is internally inconsistent. That insert on p. 57 declares that a dual sensor unbalance will result in a manual lockout (which is what I'm seeing); meanwhile, the table on p. 85 says this condition is one of those that will only cause an auto reset. The condition isn't mentioned at all later, on p.86, among those that cause the manual lockout.


Thinking about how a dual sensor can be used to detect a low water condition, the only thing I can imagine is a probe with sensors that have different response profiles. A difference between their readings due to a sudden change in temp. might be one way to detect this condition, since the hx would heat up rather rapidly under this condition. Another possibility would be if a second or third zone came on while the unit was coming up to temp, dumping cold water through the system from the new zone and causing the sensor difference to jump, if briefly. For this second case, I guess I would have thought the control software would be expecting (and accomodating ) a momentary delta-T, since it just brought a new zone on line. But maybe it doesn't. If not, I'm inclined to think this is the most likely cause. Why the unit waited 5 years for this may just be a question of aging hardware - we've had three zones since the thing was installed.


SO, thinking back on Trooper's comment about blaming the messenger, the possibility exists, of course, that the OUT1/OUT2 sensor is working fine, and legitimately detecting something the control software is programmed to see as an unbalance indicating a possible low-water condition. It's also conceivable that a momentary low-water condition or a new zone coming on might also cause the area surrounding the F1/F2 sensor to heat (or cool) rapidly as well. Since the error memory does include some instances of auto reset for "Temp rise to quickly", if these dual sensors are designed to react to a fast delta-T, then the thing may simply be dancing between the 'soft' (auto reset) and 'hard' (manual lockout) limits due to the condition mentioned by wogpa67, down below (i.e., poor flow rate). Wish s/he'd gotten back to me regarding what was wrong with the plumbing, as maybe that is indeed the issue, and the system has been marginal all along, just now showing failures because of age, etc.


Thanks very much for catching this, zoesdad - this is more to pass along to my guy when he comes by. Much appreciated.

cheers,

Ron
 
  #34  
Old 01-20-15, 03:31 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 2 Votes on 2 Posts
I believe that those dual sensors are thermally coupled closely enough that unless one of the sensors failed, they would both read the same regardless of a zone opening or closing.

The fact that one is looking at water, and one at the flue gas temp, two very loosely coupled parameters (they DO have some bearing on each other)...

No, I keep going back to what the two sensors have in common.

My answer is: they have the same connector in common.

I think something is affecting both of the dual sensors in HARDWARE... and I know how software guys like to blame the hardware! This time though, I think it's a valid blame game.



If it was mine, I would already have had the controller out on the bench, strapped on my ESD wrist strap, opened up the box, and with a bright light and an eye loupe, inspected the area around that connector.
 
  #35  
Old 01-20-15, 04:36 PM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
Trooper, FWIW, this was my initial thought as well (way back when I noted that both probes come off of the same P8 connector) - I haven't ruled that out. It's #1 on the list of things for the contractor to check (#2 is to double-check the condensate trap) and, as I say, he'll be by this week. Meanwhile, the system is functioning without problems.

All that said, I have to wonder: what's the point of a dual sensor probe if both sensors respond to temp. changes in exactly the same way? If one sensor in a dual probe goes consistently bad, the unit is going to shut down and won't restart (or, at least, won't run for very long) until it's replaced. That's a liability, not an asset. If one sensor (or its bias voltage) goes, eh, "flooby", and starts giving false indications of a low water condition, that's even worse. As such, a probe with two identical sensors would provide no practical advantage over a single-sensor probe. Right?

So I'm thinking there must be some difference between the sensors in these probes that gives them the ability to detect a 'low water' condition - especially since the notes zoesdad found state that this is their purpose (at least for the Hx OUT1/OUT2 guy).

One characteristic of a low water condition would be an abrupt change (rise) in temperature, no? If the sensors have different response profiles, that would make the probe a nominal "low water" condition detector the controller could use to respond quickly, rather than measuring the rate of change over several seconds in order to detect the same thing. It sounds like a bit of a MacGyver to me, but apparently it's enough for some jurisdictions (not ours - we're required to have the external guy). And nominal rates of temperature change would allow the sensors to remain within a couple degrees of each other, just as you noted.

If these other "Temp rise too quickly" indications in the error memory are legitimate, and I have to believe they are until proven otherwise, then they indicate a rate of delta-T, at least once-in-a-while, that's marginally out-of-spec (i.e., warranting "only" an auto reset pause). If that rate were to increase, and cause the dual sensors to go "out-of-balance", mimicking a low-water condition, then the manual lockout makes sense, again, based on the inset notes (but NOT based on the tables of error conditions on pp. 85/86, which is what I was using as a guide). And if that rate increase were caused by temperature variations from the controller's inability to deal with zones coming on, this would explain the behavior.

Again, thanks for your thoughts and time on this. I'm going to pass along your suggestion to either rule this out as the cause, or fix it.

cheers,

Ron
 
  #36  
Old 01-20-15, 05:01 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 2 Votes on 2 Posts
As such, a probe with two identical sensors would provide no practical advantage over a single-sensor probe. Right?
I think it's a way of providing a redundant readback in order for the micro to know that the probe is functioning. The more I think about it, that's the conclusion I keep coming back to. Redundant error checking.

Your theory of them having different 'curves' could easily be tested with a multimeter. I almost guarantee they are the same resistance... in fact, there may be a chart in the manual for resistance/temperature.

One characteristic of a low water condition would be an abrupt change (rise) in temperature, no?
Depending on where in the flow stream it's located, and it's proximity to the burner, it could go DOWN too... in fact, this is one of the reasons that LWCO was developed. When water gets low, and no longer contacts the aquastat bulb in a conventional boiler, it's temperature does NOT rise, and the boiler continues to fire until such time as the entire cast iron and everything is hot enough to trip the a'stat.

Low water cut offs I believe are almost always exclusively conductivity testers. Relying on a water temperature sensor is unreliable, and 'iffy' at best.

Remember, the two mediums that are being measured are very different from each other.... Flue gas and water temp. Yes, they will proportionally change with each other, but my logic keeps asking... what is truly common with the two sets of sensors... and that keeps leading me back to that control board.
 
  #37  
Old 01-20-15, 06:28 PM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
Redundant error checking.
Good thought. This was my initial assumption as well.

But the note zoesdad found proves that these dual sensors are not redundant. They're operating as differential sensors, according to the statement in the manual. Differential operation and redundant operation are pretty much mutually exclusive. For differential, you're taking both values as "true" and basing a decision on their difference - that's what this unit is doing. For redundant operation, you're looking at one and, if it looks whack, checking it against the other to decide which one to believe.

If the unit continued to function after one sensor in a probe went 'flooby', it might be arguable that these sensors provide redundant checking. But it doesn't, it shuts down. As soon as there's an 'imbalance' - that is, as soon as there's a >10º difference - the unit goes into manual lockout, according to the inset notes and based on the behavior / data I've recorded. That's not redundancy. That's using a single probe, which contains two sensors, to detect some differential condition. If that condition involves a (sudden) change in temp - up or down - then this could only be accomplished using a dual probe with sensors that have different response profiles relative to temperature change. This could be done by simply insulating one sensor differently from the other within the same probe - just wrap them in materials with different specific heat characteristics.

As for testing this theory with a (single) multimeter, you couldn't. At any given (static) temperature you wouldn't (or shouldn't) see any difference - which is exactly what you see under normal operating conditions while monitoring the 'live' temperatures on the front panel (the temps indicated are rarely more than 1º apart). In order to test this, you'd need to vary the temperature and watch BOTH resistances change in response simultaneously. You'd need TWO DMMs for that (and a way to change the temp quickly - like having the probe out and dropping it in some cold or hot water, etc.), and a dual-channel O-scope would work much better.

I completely agree that using a water temp sensor for LWCO is unreliable, and apparently so does the State of Connecticut, but that's quite clearly what the note in the manual is stating on p.57:
The U-Control module also provides low water protection by using a dual sensor to monitor the temperature of the heat exchanger. [The U-Control constantly monitors the sensor signals to ensure they are within 10 °F of one another. If the sensor difference exceeds 10 °F, the boiler is shutdown on manual reset lockout.] This method is accepted in many areas as meeting the requirement for a low water cut-off.

Some codes or jurisdictions may require additional external controls for high limit and/or low water cutoff protection. Follow the guidelines in this manual to install additional controls when required.
Now, how in the world can W-M claim that a dual sensor like this can be used to detect an LWCO condition unless the sensors within the probe are designed to react differently to sudden changes in temperature?

Anyway, that's why I referred to this as a "MacGyver", and I'm fairly certain that's why our local code requires an external (likely conductivity-based) LWCO unit in-line with the main circulator.
 
  #38  
Old 01-20-15, 07:11 PM
Member
Join Date: Jan 2008
Location: Southeastern Pennsylvania
Posts: 2,947
Received 0 Votes on 0 Posts
(Well I type so slow I see I am several posts behind, but I’ll post anyway and go back and see what you guys said, guess that’s backwards. But that’s me lol)

Sure thing Ron! Happy if I helped any at all. I too noticed that the older manual was inconsistent with respect to auto vs. manual reset for the diffs. (We used to have professional tech writers who would scrub manuals like these squeaky clean, they were great at it. But many of our manuals were for very critical military systems; commercial seems so much sloppier-to me anyway, lol).

Boy this is a bizarre problem.

I was scratching my head trying to imagine a firmware bug in which the firmware enters a condition where only one, not both, readings of the dual-sensor are taken and thus just one reading is periodically updated in the data structure for the sensor-pair. So the spread for a given sensor-pair in the data structure increases over time and is finally recognized by the watchdog as exceeding the number 10 and the error handling code is then activated. But that would have to happen for both the F and O sensors at the same time (oh boy!).

And it’s hard to imagine any intervening logic between the reads and updates for the given sensor-pair anyway? Hard to imagine how you could ever read one of the pair – but make a decision not to read the other in the pair. So therefore that means both sensors must be failing at the same time, lol – but wait a minute, wasn’t that the start of the entire discussion? lol

Maybe your hunch has been correct all along, a bias voltage problem, or maybe something in a common multiplexed path as Trooper mentioned. Ah shucks – I give up!

I’ll keep an eye out and if I see any posts anywhere about this problem I’ll post back.

Good luck!
 
  #39  
Old 01-20-15, 09:02 PM
lawrosa's Avatar
Super Moderator
Join Date: Dec 2010
Location: Galivants Ferry SC USA
Posts: 17,806
Received 14 Votes on 12 Posts
So hopefully you can describe what's incorrect about it, so I can point that out to my guy when he comes by this week (he didn't plumb it, FWIW). If something needs fixing, we can consider adding that to the DHW system quote.
I do see the circs on the return on the system. a bit too close to the tees IMO. Should be on feed in this application...

Does it have bearing??? Possibly..

Just to ask has the heat exchanger been cleaned? Most techs dont do this procedure..

Check all connections..

Name:  0117141003-00.jpg
Views: 1598
Size:  44.8 KB


The original installer did not add the sentenial 100 water conditioner.. Maybe why the other poster gets holes in his exchangers..

Name:  0117141004-01.jpg
Views: 1518
Size:  23.2 KB

Condensate was plugged ..

Name:  0117141245-00.jpg
Views: 1572
Size:  26.4 KB

heat exchanger was plugged too. Had not been serviced properly in the 4 years it was installed... All guy did was spray Windex on the outside jacket and make it shiny..

Name:  0117141228-00.jpg
Views: 2266
Size:  35.5 KB

Also pressure needs to be higher on these boilers... 15- 17 psi?

Name:  0117141146-00.jpg
Views: 1567
Size:  24.0 KB


Any bearing on the issue? Dont know...
 
  #40  
Old 01-26-15, 03:00 PM
Member
Thread Starter
Join Date: Jan 2015
Location: U.S.
Posts: 15
Received 0 Votes on 0 Posts
Thumbs up

So, just to put a temporary punctuation mark on this, let me again thank those who've kicked in suggestions - much appreciated. System's still operating fine at the slightly higher ambient temp. Clearly, whatever the problem was, it's associated with coming up to working temp. from a cold start.

My service contractor came out to size up the system for addition of an indirect DHW unit. He checked everything discussed here and all was nominal. He did NOT open the Hx because that was serviced recently, and I witnessed that being disassembled, cleaned and reassembled. No Windex. ;-)

We discussed the recent behavior and he concurred, FWIW, with the idea that the only way the probe in the Hx could act as a LWCO detector was if its dual sensors had different response rates (i.e., to changes in temp.). His experience is that these systems - especially when installed by old school plumbers, etc., which ours was - can take from 5 to 10 years of operation before beginning to show these kinds of symptoms if anything was borderline from the start. That said, he's not a W-M designer, so he's guessing just like me on the sensors, but at least the guess is based on a couple decades' experience with these systems rather than just technical gut instinct.

Thanks again,

Ron
 
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: