How, exactly does

Harvey Ramer

Auto Adapt technology work. I asked the Grundfos Rep and he couldn't explain. I have tried it but didn't get very desirable results.



I guess if I knew exactly how it worked I would know what applications to use it in.

I have no doubt it works.

 I just want to know how!



Harvey

bill_105

I'll second the question

I've put in several of Alphas and told the HO how great the thing is. It's a little awkward to not have a decent answer for such a neat gizmo. But they love the LED numbers:)

Derheatmeister

Delta P...

Electronically.... by checking the Piping Resistance"Head pressure" that needs be able to overcome..

It uses the "winding/rotors resistance values to establish a base flow condition..

It takes a couple days under different flow conditions for the "CPU/Brains" to learn about the system flow characteristics..

That is why it is important to leave the power supply to it plugged in at all times...Unplugging or switching on and off will just confuse the cir. and cause it having to relearn everything it had established..

That how i understood it...

Zman

Good question

The  Grundfos guys have told me that auto adapt only works if the circ is wired always on.

I have only tried it on radiant systems and was not happy with it. If a 1or 2 loop zone called, it would not respond at all. It is a little better with the check valve removed.

 I wire the alpha to the switch and run it in constant pressure mode. It works well that way.

I imagine it would work well in a constant circ system.

As far as how it works, I am not really sure.

Carl

Harvey Ramer

I understand that

part to. But; how can it determine a baseline without some preset pressure targets or an on-time measurement. I can understand how it could work in a TRV system where you have floating flow characteristics directly related to heat demand. But take for example a zoned system with a wide range of flow and head loss across the various zones. How will the pump interpret that? Or a high-mass radiant system combined with ODR. How will it know?



Harvey

Harvey Ramer

I wish

Grundfos would swing by here on the Wall and give us all the technical details. We are not idiots. We would understand.



Harvey

Derheatmeister

It never sleeps..

The Circulating pump is>>>> always

Derheatmeister

It never sleeps..

The Circulating Pump is >>>> Always

Mark Eatherton

A White paper on the subject from the manufacturer….

http://www.heatinghelp.com/files/posts/2461/AutoAdapt-White-Paper.pdf



Good luck interpreting…



ME

Harvey Ramer

This is how I understand it.

Correct me if I'm wrong.



The pump come with a preset proportional-pressure curve. It operates on this curve for a period of "time". After the preset time has elapsed the pump looks at the highest system pressure drop and the lowest system pressure drop and establishes a reference line in the center of the low and high pressure. It then looks at the total period of "time" and where the system pressures spent the majority of the time, above the reference line or below the reference line. The pump wants to see the system pressures spend an equal amount of time above the reference line as below the reference line for the preset period of "time".  If the system pressure spent the majority of the time above the reference line, the pump would determine that the proportional-pressure curve is set to low and recalculate a new curve to try and achieve a balanced reference line.



My opinion on the systems it would work well with and ones it wouldn't.



As is described in the White Paper, the Auto Adapt logic works well with TRVs and a constant water temp.



I think one would see mediocre results with zone valves and a constant water temp.



I think any ODR theory flies right in the face of the Auto Adapt logic. The two are both trying to achieve the same results through totally different vehicles and can never work well together.



That's my take on it.

Thanks for the paper Mark, and let me know if my analysis is correct.



Harvey

Mike Kusiak_2

Pump Algorithm

Harvey, your analysis makes a lot of sense. It really appears that the pump control algorithm is optimized for operation with TRV's.



The way I read it, the algorithm is trying to maintain the TRV's in the center of their modulating range, regardless of the required heat load. By determining the maximum and minimum pressure differentials and then setting the curve at a point in the middle, the TRV's control range is optimized for any given load condition.



This makes a lot of sense for operation with constant temp water, but with ODR there really seems to be no benefit. If the ODR curve is set up correctly, the TRV,s will be operating almost fully open at all times. Since there will be no pressure changes, the algorithm will set the pump at a fixed point on its curve and it will remain there.

So while the algorithm will work with ODR, there will be no benefit over using a regular "dumb" pump.

Mark Eatherton

Beginning to wonder...

I just went downstairs to my boiler room to see what my smart red pump was up to. It shows 45 watts and 4 GPM. It ALWAYS shows that. It's set up for constant circ below 70 degrees F OSA .



Just as a check, I closed an isolation valve. The watts dropped to 35, and the GPM dropped to 3... The valve was COMPLETELY closed. My trust in those numbers is now seriously wavering...



In any case, a DCECM circulator is MUCH more efficient on wire to water power usage regardless of the settings, and like over in the EU, I think we can expect to see it as the minimum required efficiency standards in the future.



As an interesting side not, this new technology brings with it some baggage. If used in an older iron based system, the super magnets of the motor WILL attract, hold and cause the pump to lock up. Side stream water filtration and wart treatment anyone? Drop by Mechanical Hub and read about Eric's recent experience.



ME,

Harvey Ramer

Interesting subject

on the water quality, Mark. I recently attended a webinar hosted by John Barba. Bruce Marshall from Emerson Swan was the main speaker. It was all about water quality in both new and old systems. He represented Fernox products which I have been using and very pleased with the results. Fernox has a totally immersed magnetic filter which is really ideal for use in these old iron systems. There was a lot of very good information.



This issue definitely needs to be brought mainstream and talked about.



By the way, the Bumblebee actually has a screen built into it to prevent the Iron Oxides and Black Magnetite from entering the circulator cartridge and locking the rotor. Folks at Taco we're thinking ahead.



Harvey

Hilly

how do you get constant power?

How exactly can you give this pump constant power if you say, wish to wire it to a Switching relay? For instance you wish to use this delta P circ to supply a manifold of TRV supplied panel radiators? This way the pump works ideally with the fluctuating opening and closing of the various TRV's. But if you have no heat call or the indirect takes priority and shuts down the power to this pump, wouldn't all it's data be lost?



Sorry or do I have it all wrong? So would this particular Circ only be working in all its glory in a constant circulation style system?

hot_rod

plug it into a receptacle

some of the delta P circs come equipped with a cord and plug on the end. i have several different brands and they are all plugged in. Nice for service also when you can just unplug them.



Don't trust the gpm readouts on any if them



Maybe they should display amp draw and temperature instead of errant gpm numbers? What purpose does a gpm number that is up to 60% in-accurate do?

Hilly

found it

Just found this on there I/O manual… This was the part I was curious on. I guess it has something similar to the CMOS in a PC that will 'remember' settings.



"Zone panel control applications — when there is a call for heat and power has been sent to the Alpha pump, the Alpha will remember and restart from the last duty point and hydraulic mode."

SWEI

Even the amp draws are questionable

Perhaps they are measuring power output to the motor and not power input to the controller.

Paul Pollets

Alpha pumps

I use the Alpha's on every project and have never had any issues with their performance. I've tested the amp load for comparisons and found the amperage displayed to be spot on.

Derheatmeister

Kill o Watt.. In a nut shell..

On one of my close by installs i have been monitoring everything incl. the Energy consumption of a Grundfos Alpha Via a Kill O Watt.... It is far from any Lab Equipment that i have seen and work with... but close enough...:

AT APPROX. 6 GPM and 43W it consumes 1.07 Watts over a 24 Hr period..

This was a retrofit... The Temperatures(Two Temp) which are modulated via Viessmanns OD and the LLH are always on the mark and it has Zonevalves...No Problems here...

Mark Eatherton

Umm….

I think you meant 1.07 KILO watts consumed, which at our cost per KWH is around $0.10.



Not bad, considering the alternatives.



I have a Bumble Bee I plan on putting into the project in the mountains. Should be interesting :-)



Thanks Taco! And thanks to Steve Thompson for making it happen. Might have to wait until next heating season to give it a run for its money…





ME

BigRob

Pretty simple algorithm

The bottom line is they monitor pressure over time at fixed rpms, at a certain interval and store max and min pressure. They have a pressure sensor so that makes sense. They might have a lookup table to calculate flow based on diff pressure and rpm. It's unclear if there are two sensors for diff pressure or a single outlet pressure and some other time based calculation to come up with flow. Since the gpm values are so off in some cases I suspect the latter to save money. Under the right conditions the gpm reading is probably accurate. Flow is a scaler in the k equation and is probably calculated, so they log max/min pressure and split the difference and maybe multiply by the scaler, but it seems unnecessary. If you use zone valves vs. TRVs the k value graph would have distinct steps depending on the combination of valves open. After the learning period the curve is set and the pump operates normally and probably iterates to a more refined curve every couple weeks. The micro controller is probably the cheapest 8 bit unit possible with a small memory bank, so the program can't be huge.

SWEI

Pressure sensors there are not

The electronics measure power supplied to the motor and use a lookup table to infer head.  Wonder how they handle viscosity?  

BigRob

That makes sense..

..and is much cheaper and less complicated.  Maybe there is a temperature sensor on the body and they correlated the housing temp to the fluid temperature.  If you are not measuring pressure directly, the options seem limited to compare power at a given pressure to a lookup table at a given fluid temperature.  I don't think it would be hard to beat the performance of the Grundfos or Taco smart pumps.  Their product lines seem very marketing driven in the home heating market.  There are lots of opportunities.

Eric_51

wilo

what about the wilo stratos eco 16 rfc. I'm looking at this one for a small 4 zone setup in my home. What i under stand it constantly changes based on demand.

icesailor

Wilo's:

I put one in my 6 zone house with Zone Valves. It appeared to work well. I set the speed and it was a tad slow. It didn't catch up one really cold day. I increased the speed and it caught right up.

No more of the whole force of the Taco 007 trying to drive through the smallest zone when it was calling and nothing else. I could hear the difference.

SWEI

I know Wilo has a temperature sensor

in the Stratos line, since they need it for their ∆P-T mode (which almost nobody uses.)  I was actually wondering how they would account for glycol in their estimations.