Surge Capacity vs Recirculating Sump Size

[Burton & Cherstin Sparks]

It seems like you'd want to size the sumps underneath secondary and tertiary recirculating vermifilters to be a fraction of the surge capacity, so a surge doesn't totally flush the sump.  How do you recommend sizing the volume of the sump to factor surges?

[Dean Satchell]

Its really only the primary digester that should be subject to surge, depending on the size of the outlet from the primary digesters sump into the secondary vermifilter's sump. If you make that outlet a large diameter then yes, surge capacity might be needed in the secondary vermifilter sump. But if you make that outlet say 40mm, the surge will be "held back" in the primary digester sump and released slowly. have a look at this video:

https://www.vermifilter.com/how-it-works/displacement-surge-capacity-and-equilibrium-water-level

[Burton & Cherstin Sparks]

I guess by surge I mean something like back to back showers where they use water at a rate of 100gallons/hr. The recommended recirculating pump can do 211gal/hr (800L/hr) running 20s of every minute (70gal/hr), which I'd worry isn't enough.  For a large family where some people take 30min showers, the needed surge capacity might be too big of a challenge.  Perhaps I can simplify by removing the timers for the secondary vermifilter, run it at 100% duty cycle but divide the output into 3 or 4 parallel secondary vermifilters (stacks of egg crates), and enlarge the output from the primary sump to secondary sump.  Given that we're having to do this underground to avoid freezing, I'm using long shallow sumps which could easily have multiple stacks of secondary vermifilters over the top.  The inlet to the secondary vermifilter sump would be on one side and the output would be on the other.  Then it could clean at the 211gal/hr rate (depending on lift), which should be adequate, and I wouldn't have to make the surge capacity in the primary vermifilter so big.  It might also make the pump last longer by eliminating the on/off cycles.  Thoughts?

[Dean Satchell]

Baths (especially) and washing machines also produce surge. No probs with parallel stacks over one sump, but be aware the same number of series stacks will produce a higher level of treatment. Not sure how you would feed three vermifilters from one pump... sounds problematic.

[Burton & Cherstin Sparks]

Sorry, I should have said 3 parallel for the sake of keeping the same HLR per tower, or 4 parallel for easy division (the way I'd implement it).  It seems like a 40mm (1.5in) sump to sump pipe wouldn't hold back shower flow, but maybe bath flow for a few minutes.  I'm not clear on how to figure out needed surge capacity other than sizing it to handle the max bath and know that it'll be less than that.  What is the highest HLR you would recommend for a single secondary or tertiary recirculating vermifilter tower with the surface area of one of your barrels?  I keep thinking I need to go with parallel towers to reduce HLR and keep up with long showers.

[Dean Satchell]

I haven't ever connected a bath to my wastewater flow because in my view bathwater can and should be discharged straight to the environment. The only issue is that everything needs sizing bigger to include a bath surge. If you connect your bath then sizing your surge capacity to be the bath volume is sensible and conservative. If you don't connect your bath, size the surge capacity to your washing machine volume. Have an overflow pipe at the highest surge level, unless its okay for the surge to temporarily back up.

I do not design for HLR, because I do not layer the filter media. The whole depth of media should be coarse, because recirculation takes care of treatment level. You can fine tune HLR with your pump timer, but understand that the recirculation pump should be able to run full time without swamping the vermifilter. Forget about the academic literature, the academics all follow each other assuming the media should be layered, but that's a dead end leading to waterlogged, anaerobic media and dead worms. The key is understanding that your media should be free draining and treatment level is instead dictated by recirculation frequency. This is the paradigm shift and HLR is easily "tuned" for treatment level at any point in time with your recirculation pump. 

In terms of design, stick to one pump per tower. 

[Burton & Cherstin Sparks]

Thanks for the input!  Is there benefit for the worms or treatment in having a recirculation duty cycle <100% (e.g. only running 20s of every minute)?

Given that the shower flow rate (and duration considering the larger number of people) and especially the bath volume (already plumbed in), we ran a test maintaining 5in (127mm) of surge/backup height in a primary sump with a 1.5in (40mm) outlet and a 1in (25mm) outlet.  The flow rates were about 14gal/min (53L/min) and 7gal/min (26L/min) respectively, which suggests that we would need larger sumps, or more recirculating tower/sumps in series, as you've mentioned.  Alternatively, to save space I'm wondering about using a low duty cycle pump (with float shutoff) to move effluent from the primary sump to the secondary sump to equalize the flow throughout the day, and if more surge/backup volume is needed enlarge just the primary sump.  Thoughts?  Thank you so much for your help!!!

Burton

[Dean Satchell]

Resting the vermifilter between recirculation events might provide greater aeration to the media, because as the liquid drains the pores between media particles fill with air. With a coarse media I doubt that matters much, because the media never becomes saturated with water anyway so the recirculation pump could run continuously. In my view what matters more than duty cycle is the size of the pump (quantity being recirculated) and the volume of media. What is the volume of your sump (being recirculated from/to) and what is the size (l/min) of your pump?

That 5" of surge equates to a volume that depends on area. Your actual surge capacity is the volume above the equilibrium water level of that sump and below the bottom of your primary media. Importantly, as height of surge increases you get more pressure and therefore greater flow (as you approach full surge capacity). Also, as your outlet increases you get greater flow (40mm isn't very big for wastewater). Focus on those two things to design your system to achieve the flow rate you want, rather than size of sump (irrelevant) and how many sumps in series (again irrelevant). However, understand that the more sumps that you have in series, the slower the outlet will flow in the first sump. That is, each outlet will slow the flow from the one before it. This doesn't matter if you size the first sump with sufficient surge capacity though. I think you have a problem with getting more height so you are creating volume by increasing the width of your sump?

Because water equalises through the sumps, if you pump from the first to second sumps, the water will then flow from that second sump both ways (i.e. back into the first sump). Keep it simple and enlarge just the SURGE CAPACITY of the primary sump. Enlarging the sump will achieve nothing.

Make sure you understand displacement, equalisation and surge capacity. If necessary create a model like what I did in this video.

Cheers

Dean

--
You received this message because you are subscribed to the Google Groups "Vermifilter knowledge hub" group.
To unsubscribe from this group and stop receiving emails from it, send an email to vermifilter-knowle...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/vermifilter-knowledge-hub/2d37c63c-1914-41ca-9862-3f9840448aafn%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Burton & Cherstin Sparks]

Dean,

We did a test with 4 sumps in series, each sump 2feet in diameter, and 1inch diameter piping between. Initially the first sump could sustain over 10gal/min coming in. We then measured that a 1inch of head between each successive sump could be sustained at 4gal/min.

We did further modeling based on that measured performance.  The flow out of each sump asymptotically approaches the rate coming in, and then exponentially decays back down after the flow event. Our biggest surge, causing the biggest peak flow, will likely be from a tub (not yet installed). I hear tubs can drain at over 8gal/min, and the experiment showed that a 1inch pipe between sumps alone would not prevent that. The more surface area in the primary sump the slower you'll ramp up, effectively reducing the peak flow because the tub drain time remains fixed.

Our longest steady state flow (>30min) should be a shower (currently 1.8gal/min). We would need more than 10x the barrel surface area in the primary sump to lower the peak tub drain flow closer to the shower rate. That would take you from flushing a sump in 7min to probably over 30min, but then I'd be concerned about the primary sump having such a large volume you could go anaerobic. I hesitate to go smaller than 1inch diameter piping on passive flow (not pumped) due to the risk of clogging.

I think the simplest approach may still be to do some sort of flow equalization out of the primary sump. The first image showing just a primary and secondary sump, is one of the simplest ways to enable that. You'd ensure the output of the flow equalization pump has a constant flow a little faster than the max shower rate (e.g using a valve on the output of an electric pump). That would effectually get rid of peak flow concerns.

To minimize the risk of solids buildup in the 2nd sump, here's another way to do flow equalization, which I'm leaning toward. 

I'm looking into the possibility of using air pumps in part because the flow rate would drop as the level of liquid in the primary sump drops, ensuring flow is never faster than it needs to be.

I appreciate you pushing me to ensure I'm thinking through the details. 

Burton

[Burton & Cherstin Sparks]

I forgot to mention, I think the primary sump should be 150gal or so, and I could still use standard 55gal drums for any sumps under the recirculating vermifilters.

[Dean Satchell]

Hi Burton,

If you get 4 gallons per minute with a 1 inch head, what flow do you get with a 10 inch head? Why a 1inch pipe between sumps? Why do you discuss surface area of the sump? None of this is making any sense to me, sorry. Vermifilter.com explains things like equilibrium water level and surge capacity. Your diagrams of "flow equalisation" to address "peak flow concerns" increase the risk of failure by depending on pumps and I don't understand your logic.

Your emergency overflow should be at the level of the bottom of your primary vermifilter basket. To increase your surge capacity, focus on increasing the volume below the bottom of your primary vermifilter basket and the equilibrium water level in your sumps, and make sure there is sufficient head between those so you have greater pressure whenever you have large surges. That is your surge capacity. Once you have sufficient capacity to handle the surge from say a tub, then focus on modeling the flow rate through the series of sumps based on the head that the surge generates. 

Cheers

Dean

[Burton & Cherstin Sparks]

Dean,

Thanks for your reply!  I thought you were suggesting throttling the output from the primary sump to the secondary sump to a rate the recirculating filters can handle while designing the primary sump to handle the needed surge volume to accommodate that.  In one of your most recent communications I thought you were suggesting 2 things to focus on to achieve that: "Importantly, as height of surge increases you get more pressure and therefore greater flow (as you approach full surge capacity). Also, as your outlet increases you get greater flow (40mm isn't very big for wastewater). Focus on those two things to design your system to achieve the flow rate you want, rather than size of sump (irrelevant) and how many sumps in series (again irrelevant)."  Accordingly, in the context of our greywater surges I did some testing with a smaller primary sump outlet pipe than suggested on your website and extrapolated using larger surge capacity in the primary sump by modeling.  Without having a reference point for the maximum rate the recirculating stacks of 55gal drums can treat or how small of a pipe between the primary and secondary sumps I could go without clogging, I'm doing a little bit of guesswork.  My learning was that because a 50gal tub drain rate is faster than a shower, it will build up a surge head faster than a shower, and hence would result in a short term flow rate through the recirculating vermifilter stacks that is faster than a shower (undesirable).  I also learned that a 1in (25mm) primary sump outlet would not throttle a shower flow rate, and could let an even faster flow rate through, so I tried to pivot on the design.  Here are the uncertainties causing me to want to pivot the design:

What is the fastest flow rate you think a 55gal drum recirculating vermifilter/sump can clean at?  Your site lists a 400L/h (1.8gal/min) recirculating pump, which is much faster than the technical literature as we've discussed.

What is the smallest diameter pipe you would feel comfortable using between the primary and secondary sumps?  I'm assuming that a passive flow pipe is more likely to clog than one with a pump on it.  To add another point of reference to the 1.5in (40mm) piping listed on your site, I spoke with another professional vermifilter engineer, and he indicated they had to increase their piping to 2in (50mm) to avoid clogging. Their design, at least the one I was looking at, has an anaerobic tank prior to their primary vermifilter and a series of horizontal flow planted bed vermifilters after. I don't know if those differences could lead to increased risk of clogging.

I realize that your series of recirculating vermifilter/sump design will flow passively in a power-outage, and that you could have an emergency overflow to subsurface discharge from the pumpout tank, but it seems like there is still a risk that people could be exposed to effluent insufficiently treated for surface discharge could be pumped to the surface from the pumpout tank when power is restored.  It seems like it would be ideal to find a way to ensure that anything that makes it to the pumpout tank has made it through the vermifilter of every stage in the design.  My father-in-law is working on a sketch of a simpler way to do that compared to my last design.  I think waiting to send you that full end-to-end treatment flow concept will be more helpful than sending you my sketch of just the change area today.

Most shower heads in the United States flow between 1.5 to 2.5gal/min.  Since we have 8 full time residents, and showers can be long and back-to-back, and we only have 1 shower, it seems prudent to me to design the recirculating vermifilter/sump stacks so that they can clean at least at a 2.5gal/min rate rather than trying to accommodate 8 long showers worth of surge in the primary sump.  Hence, 2.5gal/min would be the design flow I would pick.  My wife came up with the idea of throttling the tub flow by designing a tub drain hair catcher insert that limits that flow to less than or equal to the 2.5gal/min.  Then, rather than needing to design surge capacity for a 50gal tub, I would only need to worry about the 14-20gal laundry surge, which seems more manageable.

Let me know what of the above principles I've misunderstood from your emails.  Thank you again for all your help!

Burton

To view this discussion visit https://groups.google.com/d/msgid/vermifilter-knowledge-hub/96e8b636-b6fb-4c82-8bfd-6c0ec83f0e82n%40googlegroups.com.

[Dean Satchell]

Hi Burton,

No, I did not suggest throttling the rate between sumps. The flow rate between sumps is dictated by the quantity coming in regardless, and risking surge overflow is something to avoid.

A few general principles to go through:

1. Firstly, don't ever confuse recirculation flows with sump flows. Recirculating vermifilters don't affect sump flows at all, the input and output from a recirculating vermifilter to a sump are in equilibrium, therefore do not impact displacement, surge, or equilibrium water level, which are all sump-specific issues.
2. Recirculating vermifilters require a displacement sump system. That is the cold hard reality.
3. The sump system works independently of the recirculating vermifilters. There are no dependencies.

For a given system, level of treatment is influenced by influent quantity/time. To improve level of treatment:

1. increase the volume of your sumps (so the water is resident in the sump for longer). For a given recirculation volume (i.e. recirculation pump size), a larger sump volume means a longer resident time in the sump, thus more time between entering the system and exiting the system, thus better utilisation of the vermifilters.
2. increase recirculating vermifilter volume. 
3. Increase recirculation rate.

The recirculation rate (i.e. recirculation pump size) is less important than the other two points above. The same water volume going through more media volume provides greater overall treatment levels than going through less media volume. You can always increase recirculation flows later to tune your system and improve treatment, but capacity is a design issue and much more difficult to adjust later!

A good starting point is to calculate expected total flows, then residence time between entering the system and exiting the system, based on your current design (sump sizes). Either more vermifilters in series or larger volume vermifilter for each sump, provides greater treatment level.

Remember, your primary vermifilter also acts as a flow moderator, a bit like "throttling the tub flow by designing a tub drain hair catcher insert". First step, make sure that the flow through your primary digester is sufficient, so that in peak flows it doesn't fill with water and overflow! Second step, design your primary sump so that surge capacity is sufficient for peak flows, third step design your sump series to allow for the displacement that your average flow produces.

Primary sump

The head and volume of your primary sump are important design factors to avoid overflow. Your overflow is at the level of the bottom of your primary vermifilter basket. The volume between the bottom of your primary basket and the equilibrium water level in your sumps is your surge capacity. The greater the head (height between the bottom of your primary basket and the equilibrium water level), the more flow you get under surge conditions, via displacement.

Sump series

The size of each sump's outlet significantly affects water flow rate through your system. Larger outlets increase the flow. A 100mm outlet will generate much higher flows than a 40mm outlet. You just need to price these and make a sensible tradeoff between price and risk. 

"it seems like there is still a risk that people could be exposed to effluent insufficiently treated for surface discharge could be pumped to the surface from the pumpout tank when power is restored.  It seems like it would be ideal to find a way to ensure that anything that makes it to the pumpout tank has made it through the vermifilter of every stage in the design." The reality is that if your recirculating pumps fail, the system would simply act as a septic tank and settle out suspended solids under anaerobic conditions. The treatment would be to a level suitable for subsurface discharge and the only short-term issue with surface discharge would be odour. That in itself should be sufficient to reinstate the recirculating vermifilters!

Cheers

Dean

[Burton & Cherstin Sparks]

Dean,

Thank you for the clarifications!  Code requires us to assume a max daily flow of at least 390gal/day based on the number of bedrooms, but I wonder if we should use the next tier at 470gal/day based on number of occupants... We typically only have more than a couple hundred gal/day one day a week though.

For the current design I was looking to see if I could get away with 3 series 55gal drums and clean faster.  3×55gal=165gal, which would yield a residence time of 165gal / (470gal/day) = .35days = 8.4hrs. If I were doing a standard septic system, code here would require a minimum of a 1000gal tank, which would be a retention time of 1000gal / (470gal/day) = 2.1days.  To me it seems like a residence time based on average daily flow would be flawed unless you have a residence time of at least a day.

How do you factor residence time in assessing level of treatment for a series of recirculating vermifilter/sumps? I understand that more residence time means more treatment, but it seems like I should be able to quantify that mathematically for a series of vermifilter/sump stacks.  I keep wanting to divide the daily recirculation flow by the daily incoming flow.

When you talk about "increase recirculating vermifilter volume" as a means to improve treatment level, it seems like for a given recirculation pump rate more volume doesn't impact how much of the liquid in the sump passes through the vermifilter.   Is the goal to ensure that every drop makes it through the recirculating vermifilter enough times (what I've been assuming), or that if you mixed the effluent in a sump the average treatment improves (even though some drops may get flushed out without ever actually passing through a recirculating vermifilter)?

Thanks again for all your help!

Burton

To view this discussion visit https://groups.google.com/d/msgid/vermifilter-knowledge-hub/09519c0f-e205-4248-9e6a-ffeaeeff66e2n%40googlegroups.com.

[Dean Satchell]

Hi Burton,

You can't compare residence time in a septic tank with residence time in a vermifilter system. I can't help you with your calculations on residence time, but would conservatively suggest 4 series 55gal drums with 4 recirculating vermifilters as sufficient for safe surface discharge. I doubt you are discharging to the surface in your cold climate so you might get away with less treatment. Is your code performance based or prescription? A performance-based code would specify things like minimum BOD and minimum suspended solids. Can you build and then apply for compliance based on results that prove compliance? 

More vermifilter volume means more "surface area" on the media for the bacterial slime to attach. That bacterial slime is what treats the wastewater. The worms also provide some treatment but their role is more as caretakers of the system. They keep the media porous and graze the bacterial slime to keep it under control, like cattle grazing grass. So if you have more media volume, every drop that passes through the media contacts a greater surface area as it passes through, and thus is subjected to a higher level of treatment. So it's BOTH the number of times it recirculates AND the level of contact it has with the bacterial slime each time it recirculates that dictates level of treatment.

cheers

Dean

[Burton & Cherstin Sparks]

Dean,

While we have to design for subsurface distribution in winter months, we do plan to surface discharge in the warmer months.  Our area is more prescription based (e.g. septic tank size) for domestic waste.  We requested (and received approval) to discharge to the surface as greywater, which does require a fecal coliform level of 200 cfu/100 mL or less.  In our permit application we said we'd add more stages if needed to meet that.  I'd like to target the unrestricted daytime irrigation thresholds for the reuse of treated non-domestic wastewater for direct human consumption food crops, which include BOD <= 10mg/L, TSS <= 5mg/L, and TDS <= 480mG/L among other things.

Sorry, maybe I took this comment too literally when I mentioned the septic tank size as a reference point for coming up with total sump volume across all vermifilter sumps: "the system would simply act as a septic tank and settle out suspended solids under anaerobic conditions.  The treatment would be to a level suitable for subsurface discharge and the only short-term issue with surface discharge would be odour."  When you mentioned 4 series 55gal drums [as sumps], I'm assuming that corresponds with my comment that we typically only have a couple hundred gal/day.  Should I, or the state, be concerned that the 55gal x 4 = 220gal is less than the code design parameter of 390gal (per day)?

Burton

To view this discussion visit https://groups.google.com/d/msgid/vermifilter-knowledge-hub/35c306ed-2773-4aea-b9c3-d2553f210355n%40googlegroups.com.

[Dean Satchell]

Hi Burton,

I wouldn't be concerned if 390 gallons were flowing daily through 220 gallons of sumps. I'd be much less concerned about the size of the sumps than the size of the vermifilters and the level of recirculation.

Cheers

Dean