Disagree with Fleck injector sizing

[Dean Smart]

System:
16x65" tank
3.75 CF Dowex HCR S/S, nominal 120,o00 grain
75 PSI regulated
10 grain average hardness, no iron
Supplied trim: #1 white injector, BLFC=0.25, DLFC=7.0
Target regeneration is:
-8 pb/CF NaCl.
-10 gallons brine fill/draw
-27,000 grains per CF regen capacity at 8 lbs NaCl/CF
-101,000 grains cycle capacity
The system is deliberately over sized!

Dowex minimum regen flow rate is .56 gpm for a 16" diameter bed. (0.4 - 4.0 gpm/ft*2)

White injector brine draw rate is .36 gpm. But total injector flow is 1.2 gallons per minute.

Fleck specified #1 white injector for 16", that is was distributor supplied.

BD time for #1 injector is long. DB should not change with bed capacity, only flow rates need to change. Brine draw time is 10 gal /.36 gal/min = 28 minutes. Adding a slow rinse of 60 minutes leads to a BD of 90 minutes if slow rinse is ~twice brine draw time. Dowex suggests a rinse of 3-6 bed volumes. At 3 bed volumes we have 3 *3.75 CF * 7.48 gal/CF = 84 gallons. At .84 gpm rinse the slow rinse is 84/.84 = 100 minutes for a BF=128 minutes. But B2 and RR also clear the salt.

I think that I should be using #3 yellow for [at 75PSI] brine draw rate of 0.6 gpm, 1.4 gpm rinse, total flow 2.0 gpm and BLFC=0.5; noting that BLFC flow rates must be below injector flow rates to get accurate programmed brine fill results as water flows through the injector and then BLFC.

When Fleck states in Bulletin 599 May 2008: "The treated water runs through the injector before the BLFC, therefore the BLFC must be sized smaller than the injector flow rates to be accurate in determining how much treated water is being placed in the brine tank." If they are referring to "Draw" rates, then BLFC=0.5 is to0 close Draw=0.6 and then BLFC=0.25 might be better.

--I don't know what injector flow rate they are taking about. Looking at the flow charts, is this flow "Total", "Draw" or "Rinse" when filling the brine tank? During brine draw, the drain line is open and water flows through the injector. During brine fill, the system is in service and drain line is closed and water flows through only the top half of the injector then to the BLFC. This injector flow rate is not clearly documented. In any case, a larger flow rate injector provides more BLFC safety factor VS the undefined flow rate in bulletin 599.

BTW, the diagram in Bulletin 599 is not correct. The diagram shows brine draw, not brine fill!

Is what I am suggesting OK? [I have not been able to find any sizing info dated later than 2008.]

Thanks!

 

[ditttohead]

The point of the bulletin was primarily to explain the need for the injector and refill rates to not conflict. Many assemblers were mistakenly installing larger BLFC buttons with smaller injectors, this would cause the brine tank refill rate to controlled by the injector instead of the refill flow button causing under salting.

The BW2 and RR are not used for clearing salt from the system, the BR (brine and slow rinse) cycle is for that.

It does not need to be made to complex. Simply calculate the draw time and X4, this will get you close enough. Slower brine draws tend to increase efficiency slightly as the increased contact time allows for a better exchange.

If you are drawing 10 gallons at approximately .36, your draw time is 28 minute, x 4 = 112 minutes.

The BW2 and RR would remain the same.

The refill is regulated by a pressure adjusting flow control button. If you have a .25 BLFC, the refill on your system should be set to 40 minutes. Long refill times are not bad since the system is in service during the refill cycle.

Hope this helps.

 

[Sponsor]

 

[Dean Smart]

I understand what you are saying. But a higher C system in a larger diameter tank can have the same bed depth. I see BD=60 many times in this forum. So why should a larger C system take longer? If a system regen's very slow, I would expect that the top of the bed would be deeply regen'd and the bottom would get depleted brine. So there must be a limit to that thinking. Do you think that anyone here would understand what bulletin 599's flow reference is?

 

[Dean Smart]

And thanks for your reply!

 

[ditttohead]

Maybe I am not understanding but I will try again.

In general... not always true, variances will occur... (standard disclaimer)

The brine draw is traditionally set for 60 minutes, this is based on the injector drawing the brine water out of the brine tank in 15 minutes. A larger system will typically have a larger injector. The time to displace the brine solution, including dilution calculations (estimates) is 3-4 times the draw time.

Many systems will use a smaller injector in order to increase the systems efficiency (very slight gains).

As to the bulletin, if I recall correctly the settings for that unit in the bulletin were to certain high efficiency requirements so extended brine/rinse times will typically be used if you use that chart. We usually do a simple calculation and calculate the brine/rinse cycle to 3-4 times the draw time to ensure that the end user does not get salty water and so that the system maintains good efficiency.

Keep it simple, just calculate the brine draw time, and make the BR cycle 3-4 times the draw time.

The 5800LXT is a good example of higher efficiency programming and extended BR times. It will typically program itself from 90-120+ minutes. The computer assumes you have installed the required injector size (very small) and it does the calculations automatically based on the salt settings. The SXT allows you to do whatever you want, large injector, small injector, programming it so you get salty water...

 

[Reach4]

If a system regen's very slow, I would expect that the top of the bed would be deeply regen'd and the bottom would get depleted brine. So there must be a limit to that thinking.

I think the deal is that there is a limit to how much the top of the resin can be regenerated and therefore how much sodium can be retained by the upper resin. Each pair of sodium ions that did not replace a magnesium or calcium or whatever ion will continue deeper into the bed.

 

[Dean Smart]

Bulletin 559 from from 2008 and a similar document from 2007 are wrong. Fleck has injector #3 as a part but the sizing instructions do not use #3 injector for any size of tank. If you plot injector CFM vs a tank's bed area you see an obvious error. If you change 16" #1-->#2 and 18" #2 --> #3 you can see the way things were designed before someone messed up the charts - lost in translation. The key point is that CFM/ft*2 should not change, it is the design objective!

If PENTAIR fixed this error, the result does not seem to be in the wild. My distributor has the old material.

Specific brine flow seems to be 0.33 CFM / ft*2 resin bed area. At 75 PSI total injector flow is ~3 times brine flow for a total flow though the bed of 3x 0.33 = 1CFM/ft*2 @ 75PSI . Interesting magic number. Also note that brine draw rates do not change with pressure, but slow rinse times do. So a fixed slow rinse time of 3 or 4 times brine draw time is somewhat pressure dependent.

I have created and uploaded a spreadsheet pdf that you can examine. The straight lines are a linear regression fit to the data.

Tanks 22" and 24" are not plotted as they fall off the curve because there are no injectors large enough to maintain specific flow on those larger size tanks.

In attached Bulletin 599 2008, note that the diagram on page 2 is showing brine draw, not brine fill in the BLFC context.

08/19/2015 uploaded the suggested injector sizing PDF with improved formating.

 

[Reach4]

You present a compelling case.

I see that you have CFM @75 PSI in your table heading. Did you find info as to how the injector flow varies at various PSI such as 40 PSI?

 

[Dean Smart]

Page 21 in the 7000STX service manual has flow rates VS system pressure for all of the injectors.

The brine draw rates do not vary with pressures above 40 PSI and then only for nozzles #3 and larger. All charts cut off at 20 PSI.

 

[ditttohead]

Again, no need to make it to complex. Brining is based on more than a single fixed number or equation set forth by a single resin/valve/system manufacturer. Each manufacturer will have variances and design and application differences. If I recall correctly, the chart from bulletin 599 was based on the WQA certifications which requires longer brine draw times in order to increase the contact time of the brine with the resin thereby maximizing capacity to meet the California requirements of softener efficiency. Running the brine through the resin at varying flow rates yields varying results. Obviously, programming a brine draw to be many hours long may yield a very slightly higher efficiency, but there is a point of diminishing return. Different resin manufacturers use different contact times, flow rates, dilutions, etc as well so no single statement as to the brine draw times, injector sizes etc. is correct. The best the manufacturers can do is give you the information and recommendations and you get to do the math yourself. Varying pressures, freeboards, up-flow vs. down-flow, dilution ratios, salt vs. potassium, high salting vs. low salting, etc. will all affect the cycle times.

All this being said, an extended brine/draw time is more common than shorter ones. The only negative affect will be a very slightly decreased system capacity. Assuming you run the brine/rinse cycle an extra 20 minutes with a #1 injector, the units capacity of say 3000 gallons will be dropped to 2985 gallons, basically insignificant.

 

[Dean Smart]

Dittohead: But the documentation is in error, the certification is fine, but documentation did not get QA'd

 

[proj964]

Bulletin 559 from from 2008 and a similar document from 2007 are wrong. Fleck has injector #3 as a part but the sizing instructions do not use #3 injector for any size of tank. If you plot injector CFM vs a tank's bed area you see an obvious error. If you change 16" #1-->#2 and 18" #2 --> #3 you can see the way things were designed before someone messed up the charts - lost in translation. The key point is that CFM/ft*2 should not change, it is the design objective!

If PENTAIR fixed this error, the result does not seem to be in the wild. My distributor has the old material.

Specific brine flow seems to be 0.33 CFM / ft*2 resin bed area. At 75 PSI total injector flow is ~3 times brine flow for a total flow though the bed of 3x 0.33 = 1CFM/ft*2 @ 75PSI . Interesting magic number. Also note that brine draw rates do not change with pressure, but slow rinse times do. So a fixed slow rinse time of 3 or 4 times brine draw time is somewhat pressure dependent.

I have created and uploaded a spreadsheet pdf that you can examine. The straight lines are a linear regression fit to the data.

Tanks 22" and 24" are not plotted as they fall off the curve because there are no injectors large enough to maintain specific flow on those larger size tanks.

In attached Bulletin 599 2008, note that the diagram on page 2 is showing brine draw, not brine fill in the BLFC context.

08/19/2015 uploaded the suggested injector sizing PDF with improved formating.

I am confused between CFM and GPM on your discussion and spreadsheet. The Fleck data I see on #3 injector is 0.6 GPM yet your discussion/spreadsheet is all written in CFM. Can you help out with this anomaly? Thanks.

 

[Bannerman]

http://www.convertunits.com/from/US+gallons+per+minute/to/cubic+foot/minute