I have no experience with this, don't know if it would work, and I wouldn't place any bets - but there does appear to be variable speed controls for single phase (two wire) motors.

Here's the link: http://www.anaconsystems.com/text/opti_e2.html

I heard from the Applications Engineer at Tsurumi... he says:

Recommended operating range (LB480) is 19 gpm at 13.8 psi to 56 gpm at 4.76 psi.
(It) Will not get into the 30-45 psi range with this pump.
Performance curve is attached. (2.31 Ft Head =1psi.)

Recommended operating range is 19 gpm at 13.8 psi to 56 gpm at 4.76 psi. So I am at the lower range of the recommended operating range where (2.31 Ft Head =1psi.) 13.8 * 2.31 = 32' head. That would psi at the pump, I guess and not at the heat pump, so I best just use constant flow.
About controllers:

Chris @ Anacon Systems wrote:
Upon researching the Tsurumi LB480, I came across some specs that mentioned “capacitor-start”. Our single-phase drives are NOT compatible with this type of motor. Our single-phase drives are only compatible with “shaded-pole” and “permanent split-capacitor” (PSC) type motors.

Posted By Blake Clark on 06 Jun 2012 11:34 AM
...essentially a variable speed controller alters the frequency from the standard 60hz to change the speed of the motor,

That is a lot of information for nooboo to digest! I'm workin' on it!

The Applications Engineer also said that

the pump can be controlled by a VFD, but I would not use above 60Hz, as you can overload the motor, and cause damage.

On a VFD, which way does the Hz change if the motor is slowed?

Posted By Blake Clark on 06 Jun 2012 11:40 AM http://www.anaconsystems.com/text/opti_e2.html

I spoke with their rep and their VFD is not compatible with this pump since the pump is a Capacitor-Start. Do you have a link on the pressure transducers? And how did you control them?

Yeah, that controller I linked was a long shot.... I'm really not aware of any practical way to control the speed of the the pump you've got. I see two options: 1. Go with the pump you've got and throttle it down to get the flow you want. 2. Get another pump.

Pressure transducers won't directly control a pump, they work together with specialized pump controllers to maintain a constant pressure in a system. This is the system I used, though there are others. http://us.grundfos.com/products/find-product/sqe.html

In your application, though, the only compelling reason to go variable speed vs a "right sized pump" is that I'm not sure anyone makes a suitable pump small enough. You really only need about 1/5 hp if you choose to fully utilize the discharge siphon. Problem number two with variable speed in your application is that it is a hack job to convince a pump controller to operate at pressures under about 20 - 40 psi. A fairly simple hack job, but you'd be using equipment in a way its not designed for.

Also, you gotta understand, as far as I can tell I'm the only guy who's even attempted to utilize the siphon and who spent 1000 hours matching all the components (no joke). 99.9% of installers would probably tell you to go with the pump you've got and it will work great. You can learn a lot pushing the envelope and maybe gain some not insignificant system efficiency, but, really, don't think any of this is necessary to get the benefits of geo.

Posted By Blake Clark on 06 Jun 2012 04:16 PM
two options: 1. Go with the pump you've got and throttle it down to get the flow you want. 2. Get another pump.

Forgive me, but let me try this not real world example so that I can try to understand:
Lets say we have a open loop with the highest point at the top, two vertical legs on the loop, one 30' long and one 15'. At the top of the loop at 30' above x pump there is (for the sake of argument and with round numbers) almost zero psi, and yet still have say 20 gpm. So along the 'Supply' there is a pressure range of 0, and at the pump, 15psi. Along the 'Return' there is 0 psi at the top and 15' down, there is a vacuum, I don't know, say 10 inches Hg vacuum. Is that the idea?

You're describing the setup perfectly - that's more or less what you've got. But you're not quite visualizing the pressures correctly. Think of it this way - pretend there is a tank open to atmosphere at the top of the loop. The surface of the tank is at 0 psi. On the pump leg, it's 15 psi at the pump - you're correct. So what's happening in the discharge line? Well, picture the tank up on a water tower - the discharge line will also build pressure the lower in elevation it is below the surface of the tank. So the bottom of the discharge line in this example is 7.5 PSI. Got it? If you draw it out, you can see it more clearly.

So where does the vacuum come in? Well, it doesn't exist in the "tank open to atmosphere at the top of the loop" exercise. But it does happen in a closed system, just not where you think it does. Start at the elevation of the end of the discharge pipe (or if discharging underwater, at the surface elevation) THAT'S your 0 psi mark. Now, move up in elevation from there and you start to get vacuum. At around 33 feet up, you've got near perfect vacuum, and that's as much as you can get. Any more than 33 feet doesn't get you any additional vacuum. I have a 100 foot vertical drop pipe, but only 33 feet of it contributing to the siphon.

In your example, moving up 15 feet from discharge surface, you would have roughly 12" hg vacuum at the top of the system. Moving over to the supply line, 15 feet down you're back at 0 psi. If your pump were located here it would see 0 head. Moving down another 15 feet, your pump is still only seeing roughly 7.5 psi. My conversion factors are a little off, but you get the idea. With the pump you've chosen, 30 feet of head is 20 GPM, but 15 feet is 50 GPM, aren't pumps fun?

Essentially, the head your pump is seeing is the difference between the discharge surface and the supply surface. This holds true as long as both surfaces are less than 33 feet from the top of the loop. If you go with a discharge well, those surfaces could be within a few feet of one another. If somehow the source surface were above the discharge surface, you wouldn't need a pump at all - even if your unit was higher than both of them!

And, once again, since I'm on the record for posterity - no one does it like this. One real concern is accelerated mineral precipitation under vacuum. My system pressurizes at the end of the cycle to mitigate this. More complexity, more controls. However, if you've got good water and are willing to roll the dice, you'll potentially use less pumping energy than even the closed loops!

I'm with you, Where is the YOUTUBE video? Except for the total height of 37; should be a little less, rt?; unless one is below sea level with more atmospheric pressure? The physics are pretty interesting.

Woops - yep that's right. Don't know why 37 was stuck in my brain. It's more like 33 feet. I'll go back and edit the post.

The "throttled down" method is the tried and true method for pump and dump. You should see pump wattage drop when you do this, how much, I'm not sure. The pump you've selected can't really build much pressure, so don't worry about breaking anything. It might be fun to put a gauge on the supply line upstream of the throttle to see where you're at.

At this point, I should point out a "throttle" can be anything from a manually adjusted ball valve to a flow control regulator. There are fancy ones with springs and valves and what not and there are cheap ones which are essentially a small hole drilled in something. Everyone will tell you to put the throttle on the discharge side of the units hx so that the unit's hx will be under pressure. I think that's rather silly, if only because once the pump shuts off the siphon will pull a vacuum anyway. Then there are the folks who tell you to install a vacuum break in, but I think that's even sillier because all that does is introduce air into the system potentially causing even more problems.

If in the end you want a pressurized hx at all times and avoid the complication of another set of controls for an automated cut off valve, you have to install a back pressure valve. These are not cheap at all, but if sized properly might be able to double as your flow control. Sorry - things are always more complicated than they first appear.

After reviewing your (Blake) comments on another thread, I am reconsidering my pump choice...I would like to get more details on your solution, although you have posted much already.
The GSHP I have has a controller on board with connections for a (230v) supply and in house pumps.
I chipped a hole through my basement foundation yesterday with lots of mosquitoes to keep me company. It took a few hours through 8" of well consolidated Insulated Concrete. So, I uploaded a picture of the hole with temporary 1.5" pipes...

pic

Everyone will tell you to put the throttle on the discharge side of the units hx so that the unit's hx will be under pressure. I think that's rather silly,

It's a good idea. As you mention above, keeping water under pressure decreases the amount of heat exchanger fouling (and so does only using the geo for heating). But I haven't seen data as to how significant the former effect is.

Throttles work but waste pumping energy.

In general, I agree that hx pressurization has benefits regarding fouling. If you have questionable water quality to begin with, reduced pressure could hasten mineral precipitation. I designed my system to pressurize at the end of the cycle with this in mind. In addition, (warning - this will sound absolutely crazy) I run all my domestic water through the geo's hx. When the unit is not operating for a period of time this guarantees there isn't any stagnant water. I also confirmed my well water is very low in dissolved minerals before going the uber-low pressure route. A warmed loop - which could also accelerate fouling when unit is cooling - is not a concern with an open system because the entering water temperature is stable.

If you are throttling, it DOES usually make sense to put the throttle on the discharge side of the hx - but - and I apologize if I wasn't specific - unless you also have a automatic valve installed, your system will depressurize anyway once the pump shuts off. Most open system designs probably use a valve, but some just activate the pump through the internal contacts. If you've got a valve and a throttle, by all means, put them on the discharge side.

So - to valve or not to valve. That's the first question. If you stick a dewatering pump down the hole and you're not worried about maintaining pressure when system is idle, you probably don't need to have an automatic valve on the system. Just use the internal contacts to start and stop the pump - as long as the pump is 230V! If you go with a Grundfos SQE and a CU301 controller, you'll have to install a valve. Use aux contacts in the geo unit to open and close the valve and let the pump controller control the pump. (I would guess that automatic valves are probably the number one fail-point in an open system. I spent $300 on mine, and it still screws up occasionally.)

I have not found a quality dewatering pump on the market less than 1/3 hp. If someone finds one - please post!. My broken-record argument is that 1/3 hp is WAY overkill for a high water table, unless you're installing large tonnage equipment requiring large flows. The alternative I've proposed is to take a larger, variable speed pump like Grundfos and slow it down. It is possible to run a grundfos SQE at 10 psi or so using the "resistor trick" and the standard CU301 controller and transducer. 10 psi is still 23 feet of head - measured at the top of the system - even more down at the pump. O.K. - so 10 psi is better than 40 psi, better than 25 psi, however I couldn't tell you if it is that much better than a throttled 1/3 hp dewatering pump. Probably some, but might not be worth the investment. The reason is, you still have to throttle, even at 10 PSI. If you ran 10 PSI through an open 1 1/4" pipe, you'd have a gusher!

If you want to go uber-low, sub 10 psi, you can do it, but you've got to be committed. And I mean the crazy, nut-house kind. If you go there, your pumping energy will be the envy of all the closed loop advocates :-). The goal is no throttle. 1 or 2 psi at most measured at the top of the system. That's 1 or 2 psi only if the tail end of your discharge line is at the same elevation as your geo unit. If the tail end of the discharge line is below the geo unit in elevation - like to a pond or a a discharge well - you'll actually be under vacuum at the top of the system.

You don't need a Grundfos SQE to do this, but to size a single speed pump to these specs would be challenging to say the least. Modifying an SQE and CU301 controller to pump into a vacuum/siphon (that's INTO) requires five things. 1. An automatic valve. 2. A compound pressure transducer - i.e. a transducer that can read both vacuum and pressure. 3. A prime switch. (The prime switch runs the pump at higher speed if the siphon is lost.) 4. The now infamous Radio Shack resistor. 5. A modified bladder tank used as vacuum accumulator to smooth oscillations. The control wiring is so simple I won't even explain here. Getting the accumulator tank adjusted is not quite so easy and may take considerable trial and error.

I can give you the specific part numbers and sources for all this stuff - but I'm not going to quite yet. I want to be sure I'm not setting you up for failure. You're probably looking at something close to $2,000 between pump, controller and modifications and I'd hate to have to check my rear-view mirror for crazy Alaskans.

Now, at this point, I should mention that Grundfos also makes the CU300 controller. A lot more bells and whistles than the CU301 but you have to program it with a handheld programmer - sold separately. One neat thing the CU300 lets you do is manually set the pump rpm. I almost went this route (like, at the counter ready to buy) because I thought I could eliminate the transducer-valve-accumulator-control wiring. I figured I could just fiddle with the RPMs until I got the flow I wanted and wire the remote start terminal (not included with the CU301) to the geo unit. Sounds GREAT!!!! Why the hell didn't I just do THAT???!!!! Answer: variable water table. The controller needs the feedback from the pressure transducer in order to speed up and slow down the pump automatically to account for changes in the water table. This is especially true when you've got razor thin specs to maintain constant flow without a throttle. My water table can vary as much as 60 or 70 feet seasonally or with heavy domestic use (it's a 400 foot deep well). IF you had a rock-steady water table, I'm talking mere inches variation, you can go this route and maybe save yourself a little headache - or not. You'd still need a prime switch and if you want to pressurize at the end of a cycle, you'd still need a valve.

I did some calcs - so lets do your system by the numbers. The first thing to understand is how to convert head and flow to theoretical watts. The equation is head in feet * flow in gallon per minute * 8.3 (weight of water per gallon) * 0.0225969658 (conversion from foot pounds per minute to watts).

With your water table at 30 feet, if you were to pump at 10 psi, your total head is roughly 53 feet. At 10 GPM, your theoretical watts with zero pumping inefficiencies is 100 watts. A properly sized pump would be 30+% efficient, so figure about 300 watts. 30% efficient can be achieved with a properly sized pump - with an improperly sized or severely throttled pump, that efficiency can plummet, as low as 5 or 10% - but lets say you got close with sizing and were at 20%. Now your pumping watts would be 500 watts - about what you'd expect from your dewatering pump.

But lets see how far we can go. As we discussed previously, if you can take advantage of the siphon at the discharge, you can reduce head even further. I think we figured it was possible to get you down to 15 feet of head. At 15 feet, you theoretical pumping power is only 28 watts. At 30 percent efficient, you'd be at 98 watts. At 20%, you'd still only be at 140 watts.

A good candidate for these specs is a Grundfos 22SQE-80. It's rated at 22 GPM at 80 feet head. The theoretical wattage to pump at this spec is 330 watts. This pump is rated at 900 watts, so its pumping at just over 36% efficiency. Lowering the RPM to meet the 10 GPM @ 15 feet head would result in almost no drop off in pump efficiency, though some in motor efficiency. I'm pretty confident you'd be at around 100 watts.

Graciously letting the Crazy Alaskans wisecrack pass,
when asked by a visitor what the water table is around here, I might just point at the lake and say "Right about there..."
At this site, the water table Is just right about 'there'.
At what height I place the pump is still open for discussion. The first consideration is to have a water supply and return line that will not freeze. Actually, my first concern is to make sure I have the water rights to use the resource and that I am not polluting it. This groundwater is so so tasty and cold, on a hot day, I might sing "Cool Water" by the Sons o the Pioneers. There is high Iron and High Manganese in the shallow water, which Does precipitate onto my pipes and valves. The deep water has arsenic and coals.
So, how deep for the pump? My plan is just where I think the barrier is to the lower water, which is the ancient sea bed that is maybe at 25'; also that depth is as far as I can easily put in a well.

SO the pump at 20', wells screen at 25' to 20', 8" casing.
(The surface water level is somewhat stable, varying by perhaps 4')
(water temps are at about 40f at 20' and should be stable at that depth)
Total Rise from water to the GSHP is 16'. So, 16+20+ other friction losses is close enough and I am with you on using the 53'total head.
I believe that in this open system, the discharge pipe, if properly sized, you know, not too big so there is a siphon occurring on the discharge, will subtract from the total head. Subtracting the discharge height, head should (might) be 53-16 = 37 feet total. That is not a bad place on this pump curve as at that head I am getting close to the needed design flow.

Sound OK so far?

Yep - you're really not looking too bad. Given my experience with 40 EWT and a 3 ton system, though, I think your design flow is probably on the high side. But if you go with that, your pump is a pretty good match.

Back to slowing motor options: Tom at Bardac said that finding the right three phase motor combined with a single phase to three phase converter could work as an option in one scenario...

Finding the right combination...

Posted By Blake Clark on 25 Jun 2012 02:38 PM
Yep - you're really not looking too bad. Given my experience with 40 EWT and a 3 ton system, though, I think your design flow is probably on the high side. But if you go with that, your pump is a pretty good match.

That is the consensus here: the flow is high on the high side, but is what the manuf asks for.

It brings up the question of what the ideal flow is through the source side HX. I would like to know what empirical evidence should be looked for showing the flow is too fast or slow.

having an option to adjust would be good, such as the Bardac OPTIDRIVE E2 - 115VAC 1-PHASE IN, 230V 3-PHASE OUT

I expect that there are people out there with open loop heat pumps that discharge 40' down a borehole or hill. Without a valve, this means that the water is continually vaporizing in the heat exchanger - a really bad idea.

Avoid the issue with the valve and then a vacuum break on the discharge. Keeping pressure high and temperature low will minimize fouling.