VFD Submersible vs high flow limit 4

[HolesintheGround]

A new well delivers its water at 500ft, but recovers overnight to 10ft. A VFD-controlled 7GPM/1.5hp submersible looks great to cover the range of depths and flows to supply the house at a constant 50psi. The pump is rated for 10gpm max, but at first use of water in the morning, it will face maybe 150ft TDH, and would pump at around 14gpm into the (small)pressure tank. A flow restrictor beyond the pressure tank and sensor can prevent long term over-capacity pumping.

Won't this low head situation lead to upthrust damage to the pump? Will the VFD prevent short term over pumping because the initial flow at 30Hz (1725RPM) maxes out at 14/2 or 7gpm? Is the upthrust a problem only above 10gpm, regardless of RPM, or does the limit scale with RPM? It may be unnecessary, but I wonder if anyone could comment on whether a Dole 10gpm flow restrictor at the wellhead would totally mess up the VFD as the manufacturer infers...and why?

More background: the pump literature (Goulds 7GS15) strongly recommends a Dole Flow control or other means to limit flow to 10gpm to avoid upthrust damage. But the VFD literature (Goulds BF20)warns against placing any valve or other restriction between the pump and the pressure sensor/pressure tank. I got the same warning, only stronger, from the customer service rep, with no explanation but the suggestion that the VFD would take care of everything due to its slow ramp.

 

[Artisi]

Surely the overpumping will be a momentary thing, the pressure tank will sense the rapid increase in pressure and shut the pump down.
The pressure tank should be the controlling device not a VFD, that's the function of a pressure tank which is there to maintain a set pressure on the discharge side of the unit irrespective of what is happening on the inlet side.

 

[BigInch]

500 ft to 10 ft?

Isn't that a rather high variation in water depth?

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[Artisi]

The 10 - 500 ft sounds like a very low yield bore, if you can pump the water level from 10 to 500 ft with a 1.5 hp pump the inflow to the well must be minuscule, either it is a poorly installed or developed well or in very low permeable material.

 

[BigInch]

Just made me think that you could either plug the aquifer if its relatively solid, or blow it out into the well, if it was relatively weak, with that much head on it.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[EdStainless]

My thought was low perm.
If you are using a VFD to control pressure, then you don't need any other flow/pressure controls in the system.

Yes, the upthrust limit is a function of speed. It does not scale perfectly, but over a limited range it looks like it does. If you don't want to pump over 10gpm at 60hz, then you don't want to pump over ~7.5gpm at 45hz. If you look at flow curves for various freq you should be able to draw a line (not quite vertical) that represents safe max flow.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[BigInch]

A VFD trying to control pressure over a range of 10 to 500 feet will be spinning the pump too slow at 10 ft to make any flow at all.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[HolesintheGround]

Thanks to all for the insights. Yes, the well is a frustrating one. Deep, low yield wells seem to be very common here in Placer County, CA (think Sierra foothills). We have 1.5 GPM with a test pump at 500ft, and the 560ft hole was solid granite. It was just drilled and tested, and I measured its recovery after the pump-out with (pardon the improvisation) a water bottle hanging from a 300' tape measure. It got back to 190' (my first sample) at a 1.5 GPM rate, then the flow seemed to decrease linearly with depth after that. My logic says the water must be entering at about 200', and filling a dry hole below it at a fixed 1.5gpm rate; then, as the well recovers above that inlet, the flow slows until their is no head to push it above around 10-15ft. Sounds like we tapped a relatively shallow source through a deep fissure.

The County powers-that-be require a well to (a) deliver a minumum of 1.0 gpm and (b) be pumpable at 5gpm for 4 hours. The 5 gpm requirement must be met with storage for low yield wells. The driller encouraged me to put the pump deep and take advantage of the 750 gallon storage in the hole itself. If you do the math, 750 gallons plus 1.5 gpm recovery allows 5gpm pumping for almost 4 hours.

The VFD/1.5hp pump is a good match to the to the dynamic range of 100-600ft head and 1-10gpm demand, but I think EdStainless makes this sound like a bad choice, since, say with two showers cracked open in the morning, the VFD will drive the pump flow beyond its (reduced) maximum flow limit, even at the reduced RPMs needed for the high recovery/lower head situation.

Also, as BigInch notes, this isn't a great well. We could consider hydrofracting to increase supply, but I am not aware of anyone doing that in this area. From other reading, I get the idea that cycling the well level hundreds of feet may not be a good idea either due to the possibility of (a)flushing grit out of the aquafer into the well, and (b)possibly growing stuff on the wet bore. Any comments here would be appreciated.

I am getting the idea that it probably makes more sense to look at a small single phase pump, say 300ft down, throttled to keep it at, say 2gpm, and feeding a 1000 gallon cistern. A separate pump would feed the house.

 

[Artisi]

"I am getting the idea that it probably makes more sense to look at a small single phase pump, say 300ft down, throttled to keep it at, say 2gpm, and feeding a 1000 gallon cistern. A separate pump would feed the house."

Sounds like a good suggestion, pumping to a storage tank with level control and then repumping to the house utilising a small pressure system will save you a lot of headaches in then long-run.

 

[BigInch]

Either flushing grit from the aquifer into the well and system or forcing grit to move around within the aquifer and potentially clogging part of the flow that is available now.

Try your small single phase pump idea... without a VFD first. I am not a fan of VFDs when large variations in flow are not required and pumps running at reduced head also have trouble keeping up high discharge heads (head loss is proportional to rpm^2), so at 50% rpm they can only make 25% of rated head at half rated flow... at 40% rpm => 16% Head and at 33% rpm they're down to 10% Head... not really too usefull in a deep well. Flow drops off proportionally with rpm.

As Artisi suggests, many times it is better to compensate for variable flows using tankage, unless cost or space is is an issue,... before trying a VFD. You can usually use a smaller pump to fill tanks for supply during periods of higher needs. Look at your minimum and maximum flowrate requirements and what you could get from a tank to see if you can make a good match there.

You can probably install a VFD later if you find you really WANT one, but you may have to raise the pump's level in the well, due to the reduced output head at lower rpms and you just might find you'll get back around to needing to use tanks AND the VFD, if the flow drops off too much with the reduced rpm.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[HolesintheGround]

Helpful thoughts, thanks again to Artisi and BigInch. I modeled well drawdown last night with excel and I included a cistern with pump as well as a small cistern-level-controlled pump in the well. I plotted drawdown vs time, with different demand scenarios. I tried different flow rates from the well to refill the cistern, and what jumped out at me was that there is no way to avoid draw downs of up to 200ft or so, even if refill of the cistern is set as low as 1.5 or 2 gpm. The problem is the well production starts at zero, and is only up to about 3/4 gpm at 200ft. It doesn't hit 1.5 gpm until somewhere below that. Frustrating...

Hopefully the drawdown won't cause problems, because I don't see a way around it.

Just had a thought...maybe put two level sensors in the cistern...if it stays above, say 60% then refill at 1/2 gpm. if it falls to 40%, refill at 5gpm. (I would use Dole flow restrictors and two solenoid valves.) The goal would be to keep the well in a narrow drawdown range except on infrequent high-usage days (several guests in the house). I am envisioning a pressure tank on the well pump, so the 1/2 gpm flow would be no problem... Unorthodox??

 

[Artisi]

If you are to pump to a storage tank make the system as simple as possible, I would install the well pump at what ever level is required to deliver what you consider is the maximum flow needed.

I can not see the reason for the complicated switching arrangement, set the low level in the tank to bring the pump on-line at some arbitary level and monitor the fill time. The pump will deliver what ever it can governed by discharge head from well water level to the tank level plus the friction losses etc - starting from a "full well" the flow will be at it's maximum and as the level of the well falls the head will increase and flow fall off, therefore the controlling factor will probably become the low level switching point you set in the storage tank (switch-on position for the well pump) You can probably model all the variables based on demand, storage tank size, well output at the various levels etc.

Trust this helps in your thinking.

 

[BigInch]

Right. Use 1 ON switch...

and 1 high level switch on the cistern (or high pressure switch on the pressure tank) = OFF. Don't want to overflow or keep the pump running when the pressure is high.

Shouldn't need anything else.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[thewellguy]

Hi Guys:

I am going to assume that this well is going to be used to feed just the house. While I don't have to worry about low producing wells in my area, it sounds like things are getting overly conplicated. The simplest form of household well systems is currely a pump tank and pressure switch. Most of the pump manufactures have introduced a vfd system (see Goulds or grundfos) that is close in price. They are selling them based on the constant pressure aspect. No fluctations in the shower, washing machine fills faster, ect.

I beleive on of the details that is getting missed is that the house should always be under pressure (houses are usually set at 60 psi) so regardless of water level the pump will alway be pumping against 138 psi. But, without seeing the pump curve, I don't know if you have to worry about up thrust. I do know that all of the major residential well pump manufactures have up thrust bearings build into them.

In regards to which is better, it comes down to a cost vs.need, space available, and usage. If you ave going to put is 3 showers with 6 heads each and you like to take 1 hour long showers, you better go with a LARGE tank and booster pump assuming you have the space. If you water usage is relitively "normal" you can go with a pump and standard tank or a vfd.

After the breack in period, I wouldn't think that you would have to worry about loose material washing into the well in a granite formation. However, with that amount of draw down in the well, you many have to be concerned about iron in the water oxidizing, but that can be treated with an iron filter.

Hope it helped.

 

[HolesintheGround]

Welcome to TheWellGuy. I think your 138psi was meant to be 138' of head, and for the pump option that started this discussion (Goulds 7GS15), the 10gpm max flow limit is exceeded below 380'...so the 60psi doesn't help much. I had hoped a VFD would alleviate this... but it appears it will not since it monitors pressure but ignores flow as it cranks up the RPMs. The interaction between VFD control and upthrust problems doesn't get any discussion in manufacturers literature. The obvious solution would be a flow limiting (Dole)valve, but those are forbidden with VFD's.

My solution will be a flow restricted single-phase pump that feeds a cistern while staying in the recommended operating range regardless of drawdown. From there I may use a constant pressure setup to the house.

 

[thewellguy]

Holes you are correct I ment 'of head. I just looked up the curve for a 7gs15. At its off the curve at ten plus gpm at 340 tdh. At a pumping level of 500' plus pressure or rough 640 tdh your only going to ger 2 gpm. I think you have the right idea with the cistern and booster.

 

[ThePumpGuru]

You are correct in that a pump designed to handle 500' water level will "over-Pump" and potentially damage itself at 10' water level. The answer is a Constant Pressure system.
Grundfos has released their Smart-Flo Constant pressure system. This pump runs at different speeds depending on the demand of the user and the conditions of the supply.
they have pumps ranging from 1/2 HP to 1.5 HP and have just introduced a 3 & 5 HP.

I would recommend sizing the pump for the max depth of 500'. The pump would just run at roughly 10% power to produce the same results at 50'.

There are over an hour of videos on these pumps, both informational & technical, at the following link.

http://www.earthmovement.net/grundfos/SQE/sqe.htm#VIDEO

Hope this info was helpfull.
Sincerely,
ThePumpGuru
Earth Movement Pumps & Equipment