After five years and repeated failures of some twelve (12) Wirsbo UPONOR A3020522 Motorized Valve Actuators, I decided to investigate the root cause of the failure(s).

Disassembly of the first four A3020522 assemblies all revealed the same failure. The END-Switch contacts that are supposed to close the R1-R2 (red wire) contacts have failed due to excessive arcing at the contacts during the "make" operation. There were no other problems with the MVAs, in fact they all operate perfectly under load (when mounted to the manifold) and each unit successfully fully opened the valve on the manifold. Manual operation of the flow pump indeed confirms that all zones heat normally.

The internal microswitch that failed is rated at 5A 120/240V. The problem that causes the arcing and subsequent destruction of the End-Switch contacts within the micro-switch is due to the faulty design of the Wirsbo Uponor A3030003 3-Zone Control Module (and likely the similar 4-zone A3030004 version).

The problem is that the current design uses a series of half-wave rectifier diodes to logically "gate" in the 24VAC transformer voltage to a large filter capacitor, a 100uF aluminum electrolytic. This capacitor actually presents a momentary a "short-circuit" to the End-Switch contacts when initially closed, until the capacitor is charged to approximately 33VDC. The arc that occurs on "MAKE" of the switch contacts briefly exceeds the switch contact ratings, so that when repeated over hundreds of cycles eventually degrades the switch contacts so that even when the switch is mechanically closed with the contacts touching, they do not make "electrical" contact (i.e., they read open), thus never firing the on-board relay which in turn fires the Flow Pump, and rendering the zone dead and the home w/o heat. Naturally, this always seems to fail during single-digit temperatures here in New England.

The MVA's can be fully restored to new condition by replacing the microswitch with a standard off-the-shelf sub-miniture microswitch, available from local electronics parts stores. Such as the Philmore Part No. 30-2501 from LKG Industries, Rockford, IL.

Because the root-cause of Uponor A3020522 MVA failures is a faulty UPONOR A3030003 Zone Control Module, new MVAs will continue to fail until the Zone Control Module is replaced with the newer style. Failure seems to take 1-2 years tops in this part of the U.S.

UPONOR, apparently re-designed the Zone Control Module recently (around 2003), to include an on-board relay and flow pump circuit fuse, but apparently [inadvertently] introduced a problem that did not exist on the original design. The photo shown here is apparently the old board, not what is actually in the box under the same part number today:

http://www.uponor-usa.com/%7E/media/Files/Technical%20D...9_06.aspx?sc_lang=en

This photo shows what appears to be a different board (probably the original design) than the one shipping now in the U.S. for at least the past 3-5 years.

The current Zone Control Modules are responsible for damaging their own MVAs when used in direct accordance with their installation instructions. The cost of damaged MVAs adds up quickly, not to mention the service calls, inconvenience and potential freezing damage that could occur due to the failure of heating systems. MVAs sell at a typical cost to the consumer of $55-$85 each! The company should be held liable to stand behind their product(s), and replace the apparently faulty design of their Zone Controller, as well as MVA's damaged by them.

It was only AFTER I discovered the problem that the local distributor in Watertown MA replaced this years set of 4 MVAs at no charge. However, he made no mention of the root cause being the Zone Control Module, nor did they offer to replace the Zone Control Module-probably because UPONOR does not yet realize the root-cause of the end-switch contact failures is their own Zone Control Module. As an interim measure, I have modified my controller to limit the short-circuit current to within the switch mfgr ratings. Taco may be a better source for Zone Control Modules until UPONOR addresses this problem with a corrected design.

BUYERs BEWARE

This message has been edited. Last edited by: vmf777,

My radiant floor heating system was installed in 1997. In 2004 2 of my Wirsbo (Sparco) motorized valve actuators failed and were replaced under warranty by Uponor. The plastic "fingers" that held a ring inside the part of the actuator that screws into the manifold got brittle and began breaking off. Apparently they were covered because of ongoing problems with the actuators. Has anyone heard about the problem and whether Uponor still covers the old Sparco MVAs?

I installed 7 Wirsbo MVAs in ~2001 on my system.

Two failed in ~2004 (on one, the bottom "fingers" broke off (like Robert's) On the other, the pins holding the carriage inside sheared off) and were replaced by the supply house via some 'extended' plan.. I was told at the time that this was already closed, but would be pushed through.

In 2005, another failed (end switch stuck 'on'), so I spent my $90 for a new one.

In 2007, one started chattering the end-switch, but it wasnt bad enough to replace (at least Im hoping the circ pumps dont mind having the relay buzz a bit .. hehe). I havent replaced it yet.

This year (Ill call it 2008), another one started chattering.. one of the replacements from 2004.

Thanks for that switch part number VMF ! Im gonna rebuild 3 of these now (kept the 2005 failed one).

I noticed that the 'old' MVAs have vertical printing on the body.. the replacements in 2004 and 2005 have horizontal printing.. with the 2005 being a different color of green. Could it be, with the horizontal printed units, that they've finally fixed the plastic breakage issues and now its just weak switches ? (mind you, thats alot of switching to ask of a microswitch anyhow..)

Oh, and FWIW, re: VMFs analysis of the controller possibly whacking the end-switches.. I do not use any controller here. The end switches are fed from a 24vac/100VA transformer, thru the switch and to a basic relay to control 2 circ pumps.

Just an update..

I couldnt find the part# VMF mentioned in the sources I checked. I did however find a matching part.. NTE Electronic # 54-417 (got mine from newark.com)

These ones are rated 10.1a (instead of 5a). Otherwise physically identical.

Hope it helps someone..

Dave:

I did hear back from the UPONOR Design Center, based in the Great Northwest of Canada.

Their lead engineer told me he did indeed confirm my analysis, that the 3 and 4-zone controllers are in fact responsible for damaging the end-switch contacts in the MVA. They are also taking my suggestion for eliminating the problem, by placing a current limiter in the circuit to prevent the damage.

The additional news was that there was another reason those switches in the Motorized Valve Actuators were more susceptible to damage, and that is that they are NOT hermetically sealed, whereas the ones in the Paraffin-based thermal valve actuators ARE hermetically sealed, greatly extending their life in the presence of an arc.

In any case, they are now re-designing the 3 and 4-zone controllers to eliminate the significant current spike, which should fix the problems with destruction of the end-switch contacts. Apparently UPONOR manufactures the Thermal Valve Actuators, but not the Motorized ones that also carry their name.....I think he said those are honeywell manufactured

I am waiting for my replacement controller and VAs after they get the new one into production.

Vince

I recently started having odd problems with my Wirsbo system--zones calling for heat but boiler and pumps not being activated. I searched and found this very informative discussion.

My system was installed in 2003 and has two 3 zone controllers (5 actual zones) and 7 MVAs. Two of my zones have two loops each and the installer ganged two MVAs on the output side (bottom) of the controller rather than splitting the thermostat input to two control sections and running each MVA from a separate section. I hope it's OK the way it is.

Anyway, I found that only one zone has a functioning limit switch, so until it calls for heat, nothing happens even though other valves are open. Since I have my thermostats mostly synchronized I only recently noticed the problem, so I'm not sure when the failures began.

Today I pulled one of the problem MVAs and popped off the metal cover. After releasing the springs I measured the resistance and found that when closed it varies from around 15k to 60kohms (every time I actuate the switch I get a different result) rather than sub 1 ohm like the NO side of the same switch. Clearly the contacts are damaged. Note that unlike VMF I have the older controller design--no relays or caps on board, yet my limit switches are failing.

I do electronics as a hobby, so I have no problem locating replacement switches and getting them installed...except for one problem. In my MVAs the limit switches are mounted to the metal plate with a single copper pop rivet. How do I remove the switch? How will I install a new one if I damage the pop-rivet? Will I have to disassemble the plate and motor to get at the pop rivet (and re-rivet the new switch)?

If any of you folks who have done the switch replacement could help, I would appreciate it. I hate to think about replacing $400+ worth of MVAs because of $20 worth of failed switches.

Thanks,
PackratDoug

Hi Doug:

Uponor is in the process of beta testing a new version of those controller boards-which they indeed confirmed are the cause of the damaged switches-good for them to admit the error. The design team is based in Saskatchewan, CA. They phoned me to confirm they have redesigned the circuit to limit the current and eliminate the arc on "Make" that is damaging the switches.

You should replace both controllers as well as all MVAs.

Interestingly, the WAX-Based, thermal actuators do not seem to be failing at the same rate as the motorized valve actuators, but this is due to the internal switch being hermetically sealed, so when the arc occurs, there in less deterioration at the contacts (no oxygen to carbonize the electrodes), so they will last longer. However, that said, this is really a cure of the symptom, not the cause of the problem, which is that the controller presents a momentary short circuit to the first end switch that turns ON in the entire system. Over time, all switches will eventually be destroyed.

To replace the Switch, drill out the rivet, and use a small flat head or low profile pan head 4-40 stainless steel screw. I used Loctite Blue thread lock to ensure the new switch stays put.

Get both controllers replaced as well as all of the MVA's damaged by them (i.e., every one) and IMHO, Uponor should replace them as it's a defect in design.

Best of luck.

VMF777

Ya, I sorta bend the top of that sleeve in, pry the switch off, then cut the sleeve and pull it out the back with needle nose (remove the one side spring).

I found that the hole in the switches is 3/32". I first tried 1/8" rivets.. busted one switch drilling it out, busted a 2nd tryin to seat the rivet. Ended up using a 3-32 stainless bolt/nut with some locktite at the nut end (still had to ream out the switch a little with a 3/32" bit). I found some 2mm bolts to try on my next switch though.

Im hoping these NTE switches rated for 10a (instead of 5a) might last a little longer.. I have no controller etc, so I think Im just experiencing exhausted switches.

Best of luck ! (and good to see ppl are still actually 'fixing things' these days instead of just throwing them away)

edit: Oh, I misread the post from Doug.. there was a factory PopRivet ? So far, the 2 of mine Ive done have been those electronics type 'tunnels' or 'tubes'.. flared out at each end. Hope I dont run across a rivet .. I guess a drill or a dremel would work ?

This message has been edited. Last edited by: DaveCarpentier,

Thanks guys. I spent a lot of time this weekend figuring out what was what with my system, how the controller pcb works, and what could cause the MVA failures.

I disassembled one of my malfunctioning MVAs and removed the switch. I mis-spoke--it's not a pop-rivet, just a regular old hollow type. I abused the top flare with an oversized bit and managed to pull the switch over it and then cut the rivet in half and remove the rest of it. I got some 2-56 x 1/2" and M2 x 12mm HW today to mount the replacements. Ya'll be careful with that loctite. It will eat plastic. Ask me how I know. Put it sparingly on the back of the plate (presumably where the nut will go)--too much and it might wick up the threads.

The switch looks like an el-cheapo thing. Not Cherry or Omron or Honeywell like you would expect to find on a quality control. I will be replacing them all with Omron 10A rated switches. For fun I took the switch apart, cleaned the contacts gently with a fine emory file and reassembled it. Took 5 minutes. Contact resistance now reads 8 milli-ohms like a good switch should. But I don't trust the junky things.

I plan to try a 1k 2W resistor across the end switch terminals (going out to the boiler room relay). This should reduce arcing. I may also add RC snubbers across each switch as well. I don't want to have to deal with this every few years and I'm not interested in beta testing another Wirsbo/Uponor controller.

If anyone is curious here are a couple of pictures of the switch guts:





And here is a rough diagram of the controller board (I didn't trace out all of the thermostat, motor control, and blinky light stuff, just the end switch stuff). Oh, I left out the protection fuse on the schematic--whoops.



Enjoy and good luck to all you DIY fixer types,
Doug

"If it is to be, it is up to me"

Hello,

I've read this thread and find that I've experienced several of the problems discussed here. I will elaborate with a brief history.

I built a new home in 1995 with radient floor heat in the basement, and on the main floor. The home has 13 zones, using a Polaris water heater and Wirsbo controllers, manifolds, and zone controllers.

The original zone controls were Thermal Valve Actuators (Sparco, I think). About half of these units failed within the first two years. The failure I experienced did not appear to be related to the end switches at the time - I believed the actuators simply were not opening and thereby not closing the switch. (I may have miss-diagnosed the problem).

I then heard about the motorized actuators, and replaced most of the units with these. I had better results for a while, but lately have been seeing increasing problems with these as well. (These were the UPONOR A3020522 Motorized Valve Actuators that are the topic of this thread).

I seem to have two different failures with the MVAs: 1)Physically broken parts as Robert and Dave reported, 2)MVAs that are mechanically stuck open (with end switch also closed) so the zone continuously calls for heat with the valve open. To date, I have not seen the problem with the end switch contacts corroding and not electrically closing, however that may be occurring as well and I just haven't noticed.

I am really frustrated with the reliability of this system, and would appreciate any advice anyone has with regards to finding more suitable components, or other advice.

Side note - I did purchase one of the "UPONOR" Thermal Valve Actuators to try. Does anyone know if these units are reliable?

Greg (Balta)

Hi Greg,

the new "egg shaped" thermal acutators that everyone seems to be using now seem to be much, much more reliable than the old actuator hardware out there.

I don't know if they would fit your existing manifolds or not, but if so, if those are the actuators you are trying, they are pretty good.

Wow - glad I stumbled onto this forum! I have similar problems - but have not dug so deep as many of you have (not so tecky...).

Question for all - my Zone control Modules have LED lights and some of the red lights are inop (I think). When those zones with the possible light issues are demanding heat, the yellow LEDs work but the valves will not function until another zone (with functioning red LED) demands heat - all will open at that time.

Is this the same issue with the valves as with previous bloggers - or is this a new one?!

I am ready to replace the control modules tomorrow - but now I may hold off. I had a contractor install 3 years ago (of course - he has the receipts) - another battle to fight.

Thank you all for the education and the enlightening!!!

Steve

quote:

Originally posted by trapper:When those zones with the possible light issues are demanding heat, the yellow LEDs work but the valves will not function until another zone (with functioning red LED) demands heat - all will open at that time.

DO you possibly mean that the valve *is* functioning, but will not turn on the circulator pump (and thus distribute heat), until another zone (with functioning red LED) demands heat ? That would indicate a failure in the end-switch circuit.

If you have a simple volt/ohm meter, turn the 'bad' thermostat up to make a heat call.. when that yellow light comes on, verify 24v ac across the terminals where the yellow wires for that zone go (your actuator drive power). Dont disconnect the yellow wires. Then, disconnect the matching red wires and check with the ohm function across the red wires (not the terminals on the controller). With the valve actuator driven at 24v, the red wires should present a short (0 ohms). Turn the thermostat back down, and when the yellow light turns off you should show no short across the red wires.

This would verify the operation of the end-switch inside the MVA itself.

Hi Guys,

First of all, thanks to Rob for the reply regarding reliability of the UPONOR thermal actuators!

I have an additional question, and some input regarding the need for protecting the switch contacts from damage.

In reading the first post in this series by Vince, he mentions that his switch contacts are damaged by a surge from the 100uF capacitor on the controller when the contacts are closed. I examined my controllers, and I do not see the capacitor or relay that is mentioned. I assume then that I have the "older" style controller.

Q: Does this problem exist with the older style controllers?


My problem seems to be the opposite - my relay contacts are welding closed. Comments provided in this thread have helped me to think this through, and I think I have figured out what is going on.

I looked over the schematic provided by Doug, which appears very similar to my system. (Thanks Doug for the schematic!!)One difference is that my power supply on the boiler side of the circuit is rectified to provide DC, since I had a lot of 60hz "buzzing" in the relay (very annoying). I built the DC supply myself using a simple bridge rectifier and capacitor, which solved the buzzing problem very nicely. However, I had forgotten a fundamental problem with DC circuits and coils.

When a coil is energized with DC and the circuit is suddenly opened (as when a switch is opened), the collapsing field produces a large current spike. This spike often will create an arc across a switch or relay contact that can eventually weld the contacts closed. Most relays that are designed to be driven using DC incorporate a diode across the coil to suppress this current spike, and protect the equipment that is driving it. I overlooked including the diode suppressor across the relay coil in my system, and this is probably why I am seeing problems with my switches failing.

It was only after reading the contributions to this thread that my problem became apparent. Thanks to everyone - this site is really helpful.

I hope that this will help others if they have problems with their switch contacts "welding" rather than corroding.


Greg.

I have a complex 20 valve/loop system, installed in 1994. Works flawlessly except for those Wirsbo /Uponor unreliable zone valves.

Original valves were thermal (Italian source) yielding approx 50% failure rate -- which could have been avoided by simple internal wire routing -- since all failed due to wire fatigue breakage at solder joint to heater (resistor) element.

Wirsbo made parts replacement with motorized A3020522's made by (US source)Sparco. These have an approx 20% failure due to "finger breakage" & "stuck in open" position.

Current Uponor MVA's A3020522's are made by Honeywell, which bought Sparco & look almost same. Uponor's manufacturer's rep says I'm a couple of years too late for free replacement.

Current Uponor Thermal VA's A3010522 are supposedly German(?) source. They cost less, draw less current -- I plan to give them a shot.

We skipped the Wirsbo zone control Modules,luckily avoiding capacitor problem.

It's amazing that these large companies put out untested product. Life testing a VA would not require expensive tooling -- or time.

A big compliment to this site, witch offsets the blizzard of advertising baloney.