Katalox Light Fe/Mn Removal System Backsliding

[Tom McDonald]

Hello,

I’ve been lurking on these forums for a while as I’ve dealt with some water issues in a house we bought in the fall of 2018. A lot of good information here helped put me on a path to a solution, but a recent water test makes me think I’m backsliding.

The property was new construction, with a relatively deep well (~350 ft). We knew that there was an abundance of iron & manganese (2.2 ppm and 0.08 ppm, respectively) prior to purchasing, but failed to realize the large impact of that iron on things like fixtures, laundry, and taste/odor as well as the difficulty in dealing with it. I sought a number of professional opinions, but most were only interested in installing a water softener setup. We also have soft water (hardness of 30). My research here and elsewhere suggested that a purpose-built system would be more effective in the long term and would be easier to maintain, so I set about getting that installed.

The setup (all with 1” Cu piping):

• Well Pump to a Pressure tank
• Pressure tank to a three valve bypass loop
• Bypass loop (treatment leg) to Fleck 5810 XTR2 + 54” x 10” media tank, filled with 1.5 cu ft of Katalox light
• Media tank to Big Blue 20” filter housing equipped with a polypropylene filter (3M Aqua-Pure AP810-2, 5 um) that is changed quarterly
• Sediment filter back to the bypass loop

The well pump will yield 8-10 gpm, and there is a 7 gpm DLFC installed. Initially, I was running a quantity-driven (after 2000 gallons drawn – close to 1.5 weeks between) backwash for 30 min and rebed for 10 min. I quickly realized that this was inadequate, as the downstream sediment filter was fouling quickly – like every 30 days. I adjusted to a every 3 day backwash and rebed for 10 and 5 minutes respectively.

After the system was running for a few weeks (February of 2019), I took some water samples in for a full panel with a state-certified water testing facility. Generally, the treatment was effective, as the iron and manganese levels were below the limit of detection. However, the pH spiked from 7.04 to 10.6 and the hardness from 30 to 114. Again, based on the experiences around here, I figured this was normal, and did more frequent flushes and monitored the pH until it got south of 8.5.

Towards the end of 2019, the aroma / taste of iron started cropping back up periodically, so earlier this month I took a sample in for a full panel test. While the iron has creeped back up (new test at 0.094 ppm vs ND after system install vs 2.17 untreated) the manganese has exploded (new test at 0.407 ppm vs ND after system install vs 0.08 untreated). For what its worth, the pH has now come all the way down to 7.6 and the hardness to 64. Now, I haven’t tested the untreated water again to look for changes in the aquafer (will do so), but I am more concerned about the inability of the filter to remove manganese. Even more worrisome is if the untreated water is not higher in manganese than before, meaning the filter could be leaching MnO into the treated water. I think that this is unlikely, but am concerned.

So, as I think about how to tackle this, a few options come to mind:

1. Add a chemical oxidant directly to the filter during regeneration via stenner pump – can’t seem to find a lot of information on this approach for whatever reason
2. Add chemical oxidant via stenner pump either via a pre-tank mixer, or into a baffled contact tank with bottom drain – seems straightforward enough
3. Air injection system – not sure how this actually works; have seen suggestions that an air gap in the media tank accomplishes this, but it seems like any airgap in a water pressurized tank would be vanishingly small and have minimal contact with the water. Seems like they would foul in odd ways too.

Am I thinking about this correctly? Are there other approaches that would be more appropriate, given some of the symptoms? I’m leaning towards to option two with a baffled contact tank, peroxide injection via stenner pump, and possibly adding a GAC filter downstream as well.

Any advice that you all might have would be highly appreciated. Sorry for the rambling post.

Thanks!

 

[Reach4]

Interesting results.

I think you don't have a softener.

1. Add a chemical oxidant directly to the filter during regeneration via stenner pump – can’t seem to find a lot of information on this approach for whatever reason

I am not a pro.

The way I think you would do this is to program the valve to add a brine draw time, as if it were a softener. You would have an injector. Let's say #00 violet, which draws about 0.17 gpm. After backwash, have a BD time of 8 minutes.

The 15 gallon solution tank would be like smaller screw-top version of a brine tank. When you fill that, add one gallon of bleach (or other oxidant), and top up with soft water.

Regen every 3 days, and refill the solution tank every 33 days.

This is approximately what my backwashing H2S+iron filter does, except I have a different media. Plus my raw water iron level is a lot less than yours. I put my system in primarily for a light H2S smell.

An improvement on this method would be to have a brine fill of maybe 10 seconds to purge the bleach out of the brine line. My controller only resolves to 1 minute, so that would dilute the solution if I did that. The valve and line are surviving having the bleach solution stay, so far anyway.

Add chemical oxidant via stenner pump either via a pre-tank mixer, or into a baffled contact tank with bottom drain – seems straightforward enough

This would be done under control of the softener or the iron filter controller, which sends a pulse to the proportional pump to meter the solution in.

I think you would want to carefully measure residual H2O2 or chlorine to get the injection rate just right so the oxidant is exhausted by the time the treated exits the media tank. Otherwise, you would need a carbon tank to remove the residual.

If you have an unpainted tank so that you can backlight the tank and see the shadow, how much does the media expand during backwash? I am just curious.

Air injection system – not sure how this actually works; have seen suggestions that an air gap in the media tank accomplishes this, but it seems like any airgap in a water pressurized tank would be vanishingly small and have minimal contact with the water. Seems like they would foul in odd ways too.

I don't know what it takes to retrofit the AIO system.

I wonder what a batch treatment of citric acid, Iron Out, or vinegar would do. I have imagined having a tub with a pump inside as a softener resin treatment for those who remove iron with a softener. Isolate the plumbing to the system, and recirculate treatment solution. I think if the solution were heated to almost the 50C 122F limit, it could be more effective. Something along the same lines could work with other media such as KL, it seems to me. I don't know if fouling of the KL is similar to fouling of softener resin. The pump would be chosen to not be capable of pressurizing to more than the working pressure of the system, even with deadhead conditions.

 

[Sponsor]

 

[ditttohead]

Oxidant regen can be effective but the Stenner metered injection with H202 will almost certainly be enough for your application. I doubt you would need the baffled contact tank for these low levels.

 

[Tom McDonald]

I think you don't have a softener.

Correct - Sorry if I didn't make that clear. Just a backwashing KL filter + sediment filter.

The way I think you would do this is to program the valve to add a brine draw time, as if it were a softener. You would have an injector. Let's say #00 violet, which draws about 0.17 gpm. After backwash, have a BD time of 8 minutes.

The 15 gallon solution tank would be like smaller screw-top version of a brine tank. When you fill that, add one gallon of bleach (or other oxidant), and top up with soft water.

Regen every 3 days, and refill the solution tank every 33 days.

With a BD setup like this, do you still need an external pump + check valve (Stenner +)? I'm assuming there isn't a siphon effect working for you here...

If you have an unpainted tank so that you can backlight the tank and see the shadow, how much does the media expand during backwash? I am just curious.

Unfortunately, it is painted black. Would love to test out exactly what you say. I assume you're concerned about the backwash flow rate being sufficient to hit the right amount of expansion? My water does consistently register at ~ 50ºF, but 7 gpm may still be a little low.

I don't know what it takes to retrofit the AIO system.

I wonder what a batch treatment of citric acid, Iron Out, or vinegar would do. I have imagined having a tub with a pump inside as a softener resin treatment for those who remove iron with a softener. Isolate the plumbing to the system, and recirculate treatment solution. I think if the solution were heated to almost the 50C 122F limit, it could be more effective. Something along the same lines could work with other media such as KL, it seems to me. I don't know if fouling of the KL is similar to fouling of softener resin. The pump would be chosen to not be capable of pressurizing to more than the working pressure of the system, even with deadhead conditions.

I like this idea. With hose bibs in the right places, you could use a sump pump to do just this. The heat might be a little tricky, but an immersion heater (like a stick Sous vide heater) might work.

---

Based on the two replies above, it seems that there are two viable solutions:

1. Regen the media using a chemical oxidant in the brine draw cycle after every backwash. This would make sure the MnO is being regenerated to "full" activity every few days. This has the added benefit of sweeping the residue of the oxidant out during the fast wash / rebed process that would immediately follow it, reducing the need to perfectly dial in the amount of oxidant needed and / or preventing the need for a downstream carbon filter.
2. Add chemical oxidant (peroxide likely as there is no evidence of choliform present) prior to the media tank using a stenner pump + check valve, triggered either off the well pump or using a proportional injector (flow meter-based stenner ala https://stenner.com/products/meter-systems/proportional-injection-system). This should enable real time oxidation of the Fe/Mn species, but may result in residual oxidant that must be titrated down and should also be supported by a downstram carbon filter.

Does that sound right? I'm inclined to go big (aka, option #2) as it seems more likely to be successful, even if it is a bit more expensive and involved on the install. Am hoping that this becomes a "regular maintenance" thing and not a "reinvent every year" kind of thing.

A few questions.

• Are the all in one proportional injection valves from Stenner (referenced above) reliable? People seem quite happy with the pump + tank combos, but I haven't seen a lot of knowledge on the integrated ones. I worry about the failure of the flow meter. Seems the alternate solution is wiring (TTL / contact closure) to the well pump when it kicks on. This would seem to place the oxidant in the pressure tank, requiring more frequent draining/service there.
• Also, is there a particular pump (model, variable/fixed, flow rate range) that is preferred for household operation (150-250 GPD of water used)? A bit overwhelmed with options at the moment.
• If I opted to put a downstream carbon filter in, would a backwashing filter be advised, or a replaceable cartridge GAC filter (like the Big Blue Sediment filter already in use)?

Thanks for all the great info.

 

[Reach4]

With a BD setup like this, do you still need an external pump + check valve (Stenner +)? I'm assuming there isn't a siphon effect working for you here...

The same kind of injector that is used to draw brine in a softener would be used. It uses venturi action powered by flow from your main pump keeping the pressure tank pressurized. The flow of water to do that pumping is much less than is used for backwashing.

You may have one in your valve already, although I think you would want one that draws slowly. They are color coded.

Does that sound right? I'm inclined to go big (aka, option #2) as it seems more likely to be successful, even if it is a bit more expensive and involved on the install. Am hoping that this becomes a "regular maintenance" thing and not a "reinvent every year" kind of thing.

That sounds right to me.

Also, is there a particular pump (model, variable/fixed, flow rate range) that is preferred for household operation (150-250 GPD of water used)? A bit overwhelmed with options at the moment.

You should get somebody more knowledgeable than me involved in that selection.

Do put in boiler drain valves between stages to let you sample for H2O2 residual.

 

[ditttohead]

We used to sell a lot of the old "all in one" systems but recently Stenner has added some new products that have made it to where I barely stock the all in one systems anymore. The new systems have integrated ontrols and electronics which eliminates the old pulse boxes. The newer pumps also have electronic variable speed capabilities instead of the old mechanical clutch design which is more prone to wear over time.