DIY UV Sterilizer - WIP Build Thread

bg2311

Community Member
View Badges
Joined
May 25, 2021
Messages
71
Reaction score
178
Location
Atlanta, GA
Rating - 0%
0   0   0
Hello!

I'm starting a build thread for a DIY project that I'm working on (if this should be in another forum section, mods please let me know). The sheer cost of consumer/prosumer grade UV sterilizers was enough to motivate me to try to build my own.

Quick required safety disclaimer:
Even brief exposure to intense UV radiation is harmful to the eyes and skin as is high-current (wall voltage) electricity. Both will be utilized to some extent in this build as would any consumer grade UV sterilizer. While I have a modest amount of experience with plumbing at this point and past electrical projects I am by no means a professional and you should not attempt to do duplicate what you see here. Do your own research. Remember, safety third first. Please don't try this at home. </warning>

Goal:

With that out of the way, my goal here is to build a DIY UV sterilizer for our reef system (currently ~200 gallons total system volume between the 120g display and 80g frag tank) for well under the cost of a consumer/prosumer unit. Thus far that seems quite feasible.

Requirements:
From what I've gathered the target wavelength needed for the sterilizer to be most effective is ~254nm and this is readily available in the form of either LED strips or traditional fluorescent UV bulbs. I'm going with UV bulbs because they will be much easier to replace when needed. The UV light emitted needs to be fully shielded when operating for safety. Only the water should be exposed to it and will benefit from being in close proximity to the bulb. Quartz glass is the material of choice for UV at 254nm because it lets most of the desired wavelength through. I'm starting with a 55w UV bulb which given the proper contact time (flow rate/surface area of the quartz coil) should be enough to properly treat our 200g system. The cost of this build must be under the cost of a prosumer unit.

Design:
I'm going to try to flip the usual design inside out and put the water inside of the quartz instead of the bulb utilizing a lab grade quartz condensing coil like the one below, but properly sized to the length of the bulb in this application. At the moment I'm sourcing a custom prototype from Alibaba. This is likely to be the most expensive part of the project however we're still talking a fraction of what a properly sized off the shelf sterilizer would cost for a 200 gallon system.

Example quartz coil:

quartz coil example.jpg


The UV bulb:

uv bulb.jpg




The enclosure housing the bulb and coil will be lined with a roll of adhesive aluminum (one of the few things that reflects the desired 254nm wavelength) to maximize the exposure of the water to the light as it passes through the quartz coil from the bottom of the unit to the top (this should help any air bubbles escape). The plan is to have the coil mounted vertically. As for the wall thickness of the quartz coil, I'm thinking of starting with something over 1mm which seems to be the typical thickness of the quartz sleeves used in off the shelf units (perhaps 2-3mm) . Doing so will slightly reduce the overall transmission of the UV light but the tradeoff may be worth it for the added structural integrity of the coil to avoid breakage from internal water pressure, during shipping, and handling during maintenance (replacing the quartz coil). Being a closed loop this will not be designed to handle the full flow of the system through it from the start. I can easily up-size the quartz coil later if I desired if I want additional flow through it (might required a slightly thicker quartz coil wall).

Safety:
As a precaution I'm thinking of adding a Neptune Apex optical sensor at the bottom of the enclosure to quickly detect any liquid and shut off all power to the ballast and utility pump feeding the closed-loop unit should a leak occur. The bulb socket will be at the top of the unit so any drips are unlikely to make it to an electrical contact. There should be no moisture inside the enclosure at all (it all passes through the quartz tube that spirals around the bulb). The external enclosure is just to safely shield the UV light inside of the unit and hold the quartz coil/bulb in place. I might add a maintenance viewing port made of UV-safe viewing glass to confirm proper operation while the unit is closed/on while protecting my eyes. The high voltage UV bulb ballast will be up out of the way in a waterproof enclosure with a drip-loop plugged into a GFCI outlet.

Parts ordered so far:
Bulb: PLT pL-L55W/TUV 55 watt dual-lamp UV bulb which utilizes a 2G11 4-pin socket - $28
Socket for the bulb: 2G11 - $6
Vertical/horizontal clip-on supports for the bulb - $2
Ballast for the bulb: Advance Centium ICN-2S54-T (54w, 2 lamp ballast) - $22
Wiring (misc 120v solid core copper insulated wire)

Parts needed: (more to follow):
Enclosure for the bulb/quartz coil
Waterproof enclosure box for the UV bulb ballast
Roll of aluminum reflective sheeting for lining inside of enclosure
Quartz coiled tube ~20" long, 3/4" OD tube, ~ 2.5" across coil ID - cost TBD (sourcing from Alibaba)
3/4" grommets/silicone fittings to plumb in
3/4" quarter turn ball valve to adjust flow
GPH in-line flow meter for Apex


6/10/2021:
I'll try to add more to this post as parts arrive and I start to fit the prototype together. At the moment I'm getting quotes for quartz coils from Alibaba. This will be the most expensive part of the build but ordering in bulk makes them cheaper. If this is successful I plan to replace the quartz sleeve and bulb as would be done with any UV sterilizer after 6-12 months so I will need to order more of them once I get the size/glass strength correct for this application.



TLDR Disclaimer: Electricity and water don't mix. Radiation is bad for you, mkay. Don't try this at home.
 
Last edited:

AlexG

Valuable Member
View Badges
Joined
Jun 20, 2013
Messages
2,112
Reaction score
4,897
Location
Illinois
Rating - 0%
0   0   0
What are your plans for heat dissipation for the bulb to keep it running cool?
 
OP
OP
bg2311

bg2311

Community Member
View Badges
Joined
May 25, 2021
Messages
71
Reaction score
178
Location
Atlanta, GA
Rating - 0%
0   0   0
What are your plans for heat dissipation for the bulb to keep it running cool?
The quartz coil with the water flowing through it that surrounds the bulb should be carrying heat away from the bulb back towards the tank, sort of like a heat exchanger. I'll have to monitor the temperature inside the unit and see if it needs some form of active cooling. May also need to add this to the list of devices that turn off (the uv bulb, not the pump for it) if the tank gets above a specific temperature threshold.
 

AlexG

Valuable Member
View Badges
Joined
Jun 20, 2013
Messages
2,112
Reaction score
4,897
Location
Illinois
Rating - 0%
0   0   0
The quartz coil with the water flowing through it that surrounds the bulb should be carrying heat away from the bulb back towards the tank, sort of like a heat exchanger. I'll have to monitor the temperature inside the unit and see if it needs some form of active cooling. May also need to add this to the list of devices that turn off (the uv bulb, not the pump for it) if the tank gets above a specific temperature threshold.

While that might work having an open air space will allow heat to rise and wile the coil itself would remain cool the surrounding air can quickly heat up. An active cooling solution might be needed but as you said testing would first be needed to see if its an issue. I would also consider humidity within the UV enclosure. Making sure the UV enclosure is located in a place where humidity can not build up inside enough to impact the bulb will be important. This is a very interesting build I look forward to seeing how it comes along.
 
OP
OP
bg2311

bg2311

Community Member
View Badges
Joined
May 25, 2021
Messages
71
Reaction score
178
Location
Atlanta, GA
Rating - 0%
0   0   0
While that might work having an open air space will allow heat to rise and wile the coil itself would remain cool the surrounding air can quickly heat up. An active cooling solution might be needed but as you said testing would first be needed to see if its an issue. I would also consider humidity within the UV enclosure. Making sure the UV enclosure is located in a place where humidity can not build up inside enough to impact the bulb will be important. This is a very interesting build I look forward to seeing how it comes along.
I think ozone can be an issue with other fluorescent UV bulbs depending on the wavelength but this bulb shouldn't be producing much at all at 254nm so if It needs active cooling in the form a shielded fan at the top/intake holes towards the bottom I can look into that. Hopefully the water carrying heat out and away from the unit is sufficient. Another possible solution would be to make the external enclosure out of extruded aluminum motor housing. It can be ordered to length with heat fins and the inside is already round/hollow. That would turn the whole exterior into a giant passive radiator. I know I've see a external computer liquid cooling solution like that in the past.
 
OP
OP
bg2311

bg2311

Community Member
View Badges
Joined
May 25, 2021
Messages
71
Reaction score
178
Location
Atlanta, GA
Rating - 0%
0   0   0
I think ozone can be an issue with other fluorescent UV bulbs depending on the wavelength but this bulb shouldn't be producing much at all at 254nm so if It needs active cooling in the form a shielded fan at the top/intake holes towards the bottom I can look into that. Hopefully the water carrying heat out and away from the unit is sufficient. Another possible solution would be to make the external enclosure out of extruded aluminum motor housing. It can be ordered to length with heat fins and the inside is already round/hollow. That would turn the whole exterior into a giant passive radiator. I know I've see a external computer liquid cooling solution like that in the past.


Perhaps something like this example image below. Aluminum is a great conductor for heat. The fins would radiate the heat to the surrounding air and a fan either pointed at the enclosure or attached to it could move the hot air away from the fins. This is again if active cooling of some sort is needed. The quartz can handle the heat. I'm not sure what the operating temperature limit of the UV bulbs are.


extruded_aluminum_motor_housing.jpg



Here's the old external PC cooler made by Zalman that I was thinking of:

reserator.jpg
 

Smoke-Town

Active Member
View Badges
Joined
Apr 28, 2019
Messages
321
Reaction score
372
Rating - 0%
0   0   0
Also interested in this project. I just don't understand why in a typical uv design (not this innovative unique design), $100 worth of uv lighting and sleeves and $100 worth of pvc can sell for $600-800... what is the big technical or labor issue that makes these things triple or quadruple their build costs in the market?
 

MrMreef

Community Member
View Badges
Joined
Aug 21, 2021
Messages
94
Reaction score
96
Location
Munich
Rating - 0%
0   0   0
Also interested in this project. I just don't understand why in a typical uv design (not this innovative unique design), $100 worth of uv lighting and sleeves and $100 worth of pvc can sell for $600-800... what is the big technical or labor issue that makes these things triple or quadruple their build costs in the market?

I think it's a combo of:
- electrical safety certification (water + thin brittle glass + high voltages)
- long term UV stability of the materials (although with some units... Maybe not)
- dosage testing and operating envelope calculation

These things to wallop something onto the costs of a unit far beyond the costs of the materials.

EDIT: the Mickey Mouse "Hiiii Germs" 11w brand-named units are a beginner tax for sure
 

rwb500

Community Member
View Badges
Joined
Apr 6, 2020
Messages
75
Reaction score
61
Rating - 0%
0   0   0
this is a fascinating project. I never thought to have the water be in the quartz sleeve and the bulb be outside of it. However, I am concerned about the durability of the quartz - that quartz "tube" cracking open would be catastrophic. Do you know how strong it is?

If you had the water going through one (or more) linear quartz "tubes" with a carefully designed reflectors to "recycle" UV light that missed the water, I wonder if you could maintain efficiency and increase durability (I guess the linear tubes could be made thicker/stronger).
 

Chrisv.

2500 Club Member
View Badges
Joined
Mar 17, 2012
Messages
3,202
Reaction score
3,810
Location
United States
Rating - 0%
0   0   0
Hello!

I'm starting a build thread for a DIY project that I'm working on (if this should be in another forum section, mods please let me know). The sheer cost of consumer/prosumer grade UV sterilizers was enough to motivate me to try to build my own.

Quick required safety disclaimer:
Even brief exposure to intense UV radiation is harmful to the eyes and skin as is high-current (wall voltage) electricity. Both will be utilized to some extent in this build as would any consumer grade UV sterilizer. While I have a modest amount of experience with plumbing at this point and past electrical projects I am by no means a professional and you should not attempt to do duplicate what you see here. Do your own research. Remember, safety third first. Please don't try this at home. </warning>

Goal:

With that out of the way, my goal here is to build a DIY UV sterilizer for our reef system (currently ~200 gallons total system volume between the 120g display and 80g frag tank) for well under the cost of a consumer/prosumer unit. Thus far that seems quite feasible.

Requirements:
From what I've gathered the target wavelength needed for the sterilizer to be most effective is ~254nm and this is readily available in the form of either LED strips or traditional fluorescent UV bulbs. I'm going with UV bulbs because they will be much easier to replace when needed. The UV light emitted needs to be fully shielded when operating for safety. Only the water should be exposed to it and will benefit from being in close proximity to the bulb. Quartz glass is the material of choice for UV at 254nm because it lets most of the desired wavelength through. I'm starting with a 55w UV bulb which given the proper contact time (flow rate/surface area of the quartz coil) should be enough to properly treat our 200g system. The cost of this build must be under the cost of a prosumer unit.

Design:
I'm going to try to flip the usual design inside out and put the water inside of the quartz instead of the bulb utilizing a lab grade quartz condensing coil like the one below, but properly sized to the length of the bulb in this application. At the moment I'm sourcing a custom prototype from Alibaba. This is likely to be the most expensive part of the project however we're still talking a fraction of what a properly sized off the shelf sterilizer would cost for a 200 gallon system.

Example quartz coil:

quartz coil example.jpg


The UV bulb:

uv bulb.jpg




The enclosure housing the bulb and coil will be lined with a roll of adhesive aluminum (one of the few things that reflects the desired 254nm wavelength) to maximize the exposure of the water to the light as it passes through the quartz coil from the bottom of the unit to the top (this should help any air bubbles escape). The plan is to have the coil mounted vertically. As for the wall thickness of the quartz coil, I'm thinking of starting with something over 1mm which seems to be the typical thickness of the quartz sleeves used in off the shelf units (perhaps 2-3mm) . Doing so will slightly reduce the overall transmission of the UV light but the tradeoff may be worth it for the added structural integrity of the coil to avoid breakage from internal water pressure, during shipping, and handling during maintenance (replacing the quartz coil). Being a closed loop this will not be designed to handle the full flow of the system through it from the start. I can easily up-size the quartz coil later if I desired if I want additional flow through it (might required a slightly thicker quartz coil wall).

Safety:
As a precaution I'm thinking of adding a Neptune Apex optical sensor at the bottom of the enclosure to quickly detect any liquid and shut off all power to the ballast and utility pump feeding the closed-loop unit should a leak occur. The bulb socket will be at the top of the unit so any drips are unlikely to make it to an electrical contact. There should be no moisture inside the enclosure at all (it all passes through the quartz tube that spirals around the bulb). The external enclosure is just to safely shield the UV light inside of the unit and hold the quartz coil/bulb in place. I might add a maintenance viewing port made of UV-safe viewing glass to confirm proper operation while the unit is closed/on while protecting my eyes. The high voltage UV bulb ballast will be up out of the way in a waterproof enclosure with a drip-loop plugged into a GFCI outlet.

Parts ordered so far:
Bulb: PLT pL-L55W/TUV 55 watt dual-lamp UV bulb which utilizes a 2G11 4-pin socket - $28
Socket for the bulb: 2G11 - $6
Vertical/horizontal clip-on supports for the bulb - $2
Ballast for the bulb: Advance Centium ICN-2S54-T (54w, 2 lamp ballast) - $22
Wiring (misc 120v solid core copper insulated wire)

Parts needed: (more to follow):
Enclosure for the bulb/quartz coil
Waterproof enclosure box for the UV bulb ballast
Roll of aluminum reflective sheeting for lining inside of enclosure
Quartz coiled tube ~20" long, 3/4" OD tube, ~ 2.5" across coil ID - cost TBD (sourcing from Alibaba)
3/4" grommets/silicone fittings to plumb in
3/4" quarter turn ball valve to adjust flow
GPH in-line flow meter for Apex


6/10/2021:
I'll try to add more to this post as parts arrive and I start to fit the prototype together. At the moment I'm getting quotes for quartz coils from Alibaba. This will be the most expensive part of the build but ordering in bulk makes them cheaper. If this is successful I plan to replace the quartz sleeve and bulb as would be done with any UV sterilizer after 6-12 months so I will need to order more of them once I get the size/glass strength correct for this application.



TLDR Disclaimer: Electricity and water don't mix. Radiation is bad for you, mkay. Don't try this at home.

It's been a while since you started this thread. How did this build go? Very curious!
 
OP
OP
bg2311

bg2311

Community Member
View Badges
Joined
May 25, 2021
Messages
71
Reaction score
178
Location
Atlanta, GA
Rating - 0%
0   0   0
This got put on hold while experimenting with sourcing various quartz tubing. I got some from eBay. Ordering anything in less than a mass quantity is rather difficult on Alibaba. If you don't want a bulk order, even something pre-designed is rather expensive for 1-2 pieces. We are about to move cross-country so the tanks are currently broken down. Taking a year or so off. When we buy another house I'll start things up again and hopefully take another crack at this.
 
Back
Top