One of the drawbacks of a twin-tank softener seems to be that as soon as tank A uses the preset amount of water it will switch to tank B and immediately start regeneration. Of course that happens when you are using water which is likely during the day when you most need it. With a large tank regeneration can use a significant amount of your water capacity which may lead to loss in water pressure. However, tanks are sized to be able to supply multiple days of softening (about a week, maybe less). It seems to me that you could have the best of both worlds if after switching to tank B the regeneration in tank A was delayed for the following night. As long as you don't use up a tank's capacity in one day (which shouldn't happen if properly sized), you would always have soft water and your water pressure wouldn't drop during the day. Is it possible/desirable to configure valves this way?
Can twin-tank softener be scheduled to regenerate at night
Users who are viewing this thread
I think the Fleck 9100SXT can do that. I am not a pro. I see you want to avoid having the softener on bypass for about 1.25 hours during regen as you would have with a single tank. Flushing toilets and washing hands during bypass is not a big deal, but a shower or laundry would be.
A twin has the advantage that it is more salt-efficient, and you would be losing that.
Backwash on a softener is not all that great -- maybe 2.5 GPM for a 10 inch tank. Backwash and fast rinse might be 20 or 25% of the service flow rate, so there is still a lot of water at a good pressure available for other uses.
A twin has the advantage that it is more salt-efficient, and you would be losing that.
Backwash on a softener is not all that great -- maybe 2.5 GPM for a 10 inch tank. Backwash and fast rinse might be 20 or 25% of the service flow rate, so there is still a lot of water at a good pressure available for other uses.
Sponsor
Paid Advertisement
The reason I'm considering dual tank is the increased salt efficiency (doesn't need the 30% buffer/reserve) and to never run out of hard water, either during regeneration or if use increases significantly during the day. We often have visitors so water use can vary significantly day to day.
Our well pump is variable and can go up to 15gpm. My estimate: (220 ppm hardness / 17.1) * (6 people * 75 gals/day) * 7 days = ~40k grain at 6lb/cubic foot salt concentration. We don't have any iron and manganese is negligible (0.007 mg/L). Add an extra 30% for a single tank system and you're up to ~52k grain. I haven't looked at specific models and their specs yet, but using this (http://media.wattswater.com/F-WQ-EngineeringGuide.pdf, page 12) as a general guide I would need about 2.5 cubic ft of resin, which is a 13" tank which backflows about 9gpm. That is 60% of the flow rate for backwashing and leaves 6gpm before the pressure starts dropping. Fill a bathtub, water the plants, etc. and I think you're pushing the 6gpm. For even more efficiency you can drop to 4lb/cubic foot of salt and/or regenerate every 2 weeks instead of 1. Dropping to 4lb seems doable, but 2 weeks would double the size of the tank and backwash flow rate.
Dual tank wouldn't need the 30% reserve capacity so I could probably drop to 2.0 cubic ft of resin which would decrease the backwash flow rate. That's 7gpm from the chart which leaves 8gpm for the house, most likely during the day.
Now, you did say 2.5gpm for a 10" tank, which is half what the chart I linked to shows. Maybe I should be looking at specs of specific models, but if it's half it would be 4.5gpm or 30% of the flow rate, leaving 11.5. That's quite a bit better than the 6gpm I came up with. Maybe I misunderstood the chart? The bottom of the flow rate column says it's the backwash rate and I assumed it was gpm.
I'm also not a pro, so if any of the above doesn't make sense please feel free to correct me. Thanks!
Our well pump is variable and can go up to 15gpm. My estimate: (220 ppm hardness / 17.1) * (6 people * 75 gals/day) * 7 days = ~40k grain at 6lb/cubic foot salt concentration. We don't have any iron and manganese is negligible (0.007 mg/L). Add an extra 30% for a single tank system and you're up to ~52k grain. I haven't looked at specific models and their specs yet, but using this (http://media.wattswater.com/F-WQ-EngineeringGuide.pdf, page 12) as a general guide I would need about 2.5 cubic ft of resin, which is a 13" tank which backflows about 9gpm. That is 60% of the flow rate for backwashing and leaves 6gpm before the pressure starts dropping. Fill a bathtub, water the plants, etc. and I think you're pushing the 6gpm. For even more efficiency you can drop to 4lb/cubic foot of salt and/or regenerate every 2 weeks instead of 1. Dropping to 4lb seems doable, but 2 weeks would double the size of the tank and backwash flow rate.
Dual tank wouldn't need the 30% reserve capacity so I could probably drop to 2.0 cubic ft of resin which would decrease the backwash flow rate. That's 7gpm from the chart which leaves 8gpm for the house, most likely during the day.
Now, you did say 2.5gpm for a 10" tank, which is half what the chart I linked to shows. Maybe I should be looking at specs of specific models, but if it's half it would be 4.5gpm or 30% of the flow rate, leaving 11.5. That's quite a bit better than the 6gpm I came up with. Maybe I misunderstood the chart? The bottom of the flow rate column says it's the backwash rate and I assumed it was gpm.
I'm also not a pro, so if any of the above doesn't make sense please feel free to correct me. Thanks!
I was looking through the 9100SXT manual and it seems the following settings would be what I want:
- Control type: meter delayed
- Number of tanks: 2
- Reserve selection: any, but FS (safety factor) should work
- Safety factor: 0%
- Regeneration time: 3am or whatever makes sense
It does need a reserve buffer of maybe 10% to account for the soft water used during regen I think. This presumes that the soft water used during regen of the other is not metered -- which is what I infer from some instructions.
You don't run the irrigation water through the softener.
That 13 inch tank would backwash at 4 GPM typically . Could be 5 if the backwash water is expected to be warmer (don't know where you would draw the line, but that 5 might be for Phoenix). What page of the Watts manual led you to 9 GPM?
See figure 2 of http://www.purolite.com/Customized/...oliteProductsManagement/Resources/rid_631.pdf and shoot for about 40% expansion I think. Purolite C-100E is the resin other makers usually compare themselves to.
Where did you find your 9100SXT manual? The manual I found said
Meter Immediate systems
generally do not use a reserve volume. However, in twin
tank systems with soft-water regeneration, the reserve
capacity should be set to the volume of water used during
regeneration to prevent hard water break-through.
generally do not use a reserve volume. However, in twin
tank systems with soft-water regeneration, the reserve
capacity should be set to the volume of water used during
regeneration to prevent hard water break-through.
Also, the Fleck manual I found did not cover the SXT programming, but the choices seem to be very similar to the 5600SXT programming.
http://www.qualitywaterforless.com/v/vspfiles/downloadables/fleck/91SXTINS.pdf goes for 10%, but I think a more precise number would be warranted knowing the DLFC and injector being used.
I was thinking it would just use soft water from the full tank to regen the depleted one, but didn't consider it may not metered at that point. If I got this I would probably watch the first few regenerations to see if the meter goes down or not and adjust accordingly.
You're right, outside bibs wouldn't go through the softener. I was just thinking of what might require high flow and forgot that detail. Still, have a couple people showering, someone washing dishes and it adds up. We have three bathtubs, a shower and a large jacuzzi-like bathtub. They probably won't all be used at once, but two or three at once happens.
4 GPM would be much better. We're in Maryland and the water is probably not considered "warm". Page 12 has a Flow Rate-Softener column which says 9 for a 2.5 cu ft tank, no unit of measurement listed. I interpreted it to be GPM.
Yeah, I saw that earlier. Figure 2 says it's gpm per cubic feet and they mention 50-75% expansion. At 50% expansion and 50 degrees temperature that's about 4 gpm. 2.5 cubic feet would be 10 gpm, right?
Thanks for the help.
I found this manual: http://www.purewaterproducts.com/img/docs/manuals/SXT_Service_Manual.pdf It does have the paragraph you mentioned and the very next paragraph is about "Meter Delayed Control" and doesn't specifically address twin tank. It seems I would need a hybrid of immediate (change to new tank) and delayed (regeneration), but I don't see any indication it's possible.
IMO, they are saying 9 GPM of forward flow rate. I believe that is low for a home, but it might be appropriate for water for a STEAM BOILER -- which they talk about on page 9 . But I was confident that they would have said backwash rate if it was backwash. Ahh.... bottom of that column says
"2 Figured @ 5 gpm per square foot of bed area. FLOW RATE OF SOFTENER IS TWICE BACKWASH RATE"
I read that to say that backwash rate for a softer would be 9/2=4.5.
It says ft2 which is square ft cross section.
Whoa, they do say
During upflow backwash, the resin bed should be expanded in volume by between 50 and 75%, in order to free it from any particulate matter from the influent solution, to clear the bed of bubbles and voids, and to reclassify the resin particles as much as possible, ensuring minimum resistance to flow.
Well that's odd, since home softeners are normally designed to give 50% "freeboard" meaning that about 1/3 of the tank is left free of media. 75% expansion would blow the resin out if you don't have a top basket.
A 13 inch tank is 0.9217520112 square ft cross section. I was going to round to 0.922, but decided to paste lots of worthless digits. Pie are square. Cookie are round?
The top of the column says flow rate and the bottom says backwash rate. I had missed the footnotes which seem to distinguish between flow rate (forward, like you said) and backwash rate. I think your interpretation mostly makes sense, but then is that 9 GPM max forward flow? That seems very low for 2.5 cubic feet. For example, this page (http://www.flecksystems.com/fleck-s...rs/fleck-2510-sxt-metered-water-softener.html) shows their 2.5 cubic feet unit as having max flow 25 GPM. Usually the max will have more PSI loss, but 15 GPM should be handled alright.
Good point on square feet vs cubic feet; I misread that. The other strange thing in the Watts PDF footnote for the softener is "Figured @ 5 gpm per square foot of bed area." You just calculated a 13 inch tank has about 1 square foot and notice they also list SQ FT BED AREA at 0.92 which matches what you said. So it seems flow rate would be 5 GPM, not the 9 GPM listed or the 25 GPM at the flecksystems.com link. Something doesn't add up...
I guess all this discussion is really trying to understand these three things:
Edit: they do say 50-75%. If 1/3 freeboard is left that meets the bare minimum which may be sufficient. I wonder if closer to the middle (62.5%) would be better, but if 1/3 is the norm I'll probably stick with that.
Good point on square feet vs cubic feet; I misread that. The other strange thing in the Watts PDF footnote for the softener is "Figured @ 5 gpm per square foot of bed area." You just calculated a 13 inch tank has about 1 square foot and notice they also list SQ FT BED AREA at 0.92 which matches what you said. So it seems flow rate would be 5 GPM, not the 9 GPM listed or the 25 GPM at the flecksystems.com link. Something doesn't add up...
I guess all this discussion is really trying to understand these three things:
- What is a ballpark maximum/reasonable forward flow rate
- What is a ballpark backwash flow rate
- Can you switch to the second tank immediately and delay regeneration
Edit: they do say 50-75%. If 1/3 freeboard is left that meets the bare minimum which may be sufficient. I wonder if closer to the middle (62.5%) would be better, but if 1/3 is the norm I'll probably stick with that.Yes. The residential use can tolerate to about 1 grain of hardness , and the boiler applications need a more stringent standard.
Backwash should be done based on area, and forward rate would be more based on volume. 1. Read http://www.qualitywaterassociates.com/softeners/sizingchart.htm It says 22 GPM, and I think that is reasonable stretch. Maybe that is tolerating a bit more bleedthrough than you would prefer. It is not a hard number.
2. Backwash 4 or 4.5 GPM with 4 being common.
3. Not with the 9100 as far as I can tell. Sounds like an interesting useful feature that maybe somebody will build in. Maybe somebody has. A software-controlled design would have the potential if the marketing people let the software people do it.
This picture illustrates freeboard.
Probably someone has already thought of delaying only the regeneration with twin tank and there is a good reason not to. I doubt with my little knowledge I have come up with a great idea, but who knows...
In terms of SFR, I'm seeing recommendations all over the place. The flecksystems link I sent said 22 GPM. Several sites say not to exceed 15 GPM per square foot of bed area. This site (http://archive.wcponline.com/pdf/0304 Water Softners.pdf) says 12.5 and for 15 I should be going up to 3.0 cu ft. Who is correct?
Another interesting bit is that some states don't allow you to send softener backflow into a septic tank/field. I couldn't find anything specific to Maryland, but this is somewhat off topic.
In terms of SFR, I'm seeing recommendations all over the place. The flecksystems link I sent said 22 GPM. Several sites say not to exceed 15 GPM per square foot of bed area. This site (http://archive.wcponline.com/pdf/0304 Water Softners.pdf) says 12.5 and for 15 I should be going up to 3.0 cu ft. Who is correct?
Another interesting bit is that some states don't allow you to send softener backflow into a septic tank/field. I couldn't find anything specific to Maryland, but this is somewhat off topic.
ditttohead
Water systems designer, R&D
I skipped through most of the posts, sorry but time is difficult to find tonight.
Simply put, the regeneration will not be noticed in a residential application. The backwash time is only 5-8 minutes and the rapid rinse should be about 4-5 minutes. The backwash rate at a "normal" water temperature would be about 4 gpm for only a few minutes. Figure a normal shower only uses 2-3 gpm, faucets use 1.5, etc, you would have to have a lot of water use going on to ever notice it. When we get boards in the shop we always go into the programming to see the history. It is rare that we ever see even larger houses peak over 8 gpm.
As to the reserve, no need to add 10% for regen... this is a complex topic but to simplify it, since you are regenerating with soft water, you are gaining capacity that would otherwise be lost... a single tank system uses hard water to regenerate but that water is not taken off the capacity...
Since most systems are already set with some reserve artificially anyway, slight increases in salt settings, or raising the hardness in the computer a couple grains, this is really not an issue.
Simply put, the regeneration will not be noticed in a residential application. The backwash time is only 5-8 minutes and the rapid rinse should be about 4-5 minutes. The backwash rate at a "normal" water temperature would be about 4 gpm for only a few minutes. Figure a normal shower only uses 2-3 gpm, faucets use 1.5, etc, you would have to have a lot of water use going on to ever notice it. When we get boards in the shop we always go into the programming to see the history. It is rare that we ever see even larger houses peak over 8 gpm.
As to the reserve, no need to add 10% for regen... this is a complex topic but to simplify it, since you are regenerating with soft water, you are gaining capacity that would otherwise be lost... a single tank system uses hard water to regenerate but that water is not taken off the capacity...
Since most systems are already set with some reserve artificially anyway, slight increases in salt settings, or raising the hardness in the computer a couple grains, this is really not an issue.
Similar threads
