kenneth wolfe
New member
I like..If you was to buy a 3d printer for home/hobby use what would you suggest ?
180 gallon build: Modular control system, 3D printed equipment, open-source
- Thread starter jrhupp
- Start date
If this question was directed to me, I am afraid my answer might be a bit disappointing; I don't know enough about what is available in the market to make any sort of educated choice. I chose the kit I did (almost two years ago) based on price and reputation of the manufacture. I was budget limited which prevented me from evaluating anything that came preassembled and just don't have a ton of experience with the plug and play machines that are out there.
I can summarize what experience I do have with other machines:
- A friend owns a MakerBot. I don't know the specific generation, but it would be the previous version of whatever was the newest generation circa 18 months ago (I know real helpful). He got the newest generation when he first purchased but said it was endless problems and swapped it for the previous generation. What I have seen of it suggest high reliability and very high print quality. It seems they offer quite well developed software solutions that make it easier to get going with things.
- We run a Stratasys printer at work. It performs amazingly! If I were stupid rich or a large engineering company, it is what I would buy. But keep in mind this is not a hobby printer.
- I visited with some folks running a Luzbot printer. They loved it and what I saw from it suggested high print quality. It looked very clean and quite well made.
If you want an out of the box solution and have no experience with FMD printing, don't get the printer I did. Aside from it being a kit. There was a steep learning curve to really understand how it worked and then take that knowledge to tweak the system to get the performance/print quality where I need it to be. I get excellent performance out of it now, but it took a lot to get there.
And just so I don't give the wrong impression; I like my MakerFarm and they have provided excellent customer support when I have contacted them.
kenneth wolfe
New member
Thank you.. I have see a few diy kits and was leaning towards that.
kenneth wolfe
New member
Could I talk you into printing hybrid impeller wheel for Jebao pump?
I'm not necessarily opposed to doing this, though at first pass I don't think 3D printing is a good technology for this type of part. As I understand it, this is a part that would include a large number of small bits and be under a lot of stress during use. I'm curious to see what the part may look like though. If there were a way to adapt it to printing and have something that would be reliable, that would be right in one of my areas of interest.
Do you have a model for it? At this point, its not something I know enough about to draw up and I don't have specs to be able to fit something to a specific pump.
So progress has been slow, but I managed to fight past my cold and the miserably cold weather here (my garage is unheated) and get a few things done this weekend.
I'm still chipping away at the plumbing. The return lines are mostly done. I still have a couple of custom PVC parts I want to turn on the lathe but it is just too cold for that now. The drains are plumbed from the overflows to the sump. I still need to decide how these terminate into the sump. And I still need to drill and plumb my filter sock/rubble box to the sump. But its getting there.
I finally got my DIN rail and started to layout my equipment panel. I still need to print up a few more sections of the cable track and I don't have all of my modules built yet that will go in here. For sure, I still need to complete the serial switch (multiplexer), the main controller and the 12VDC PSU. Here is a shot of the panel as it is now:
With the door shut:
Here is the backside of the panel. Since it is exposed inside the cabinet I didn't want bolts and what not sticking out. So I used T-nuts for mounting the DIN rails; nice and clean.
For other progress points, I got the temperature sensors for my controller put together. These are built from 10kOhm@25C 0.1% tolerance thermistors potted into stainless steel tubing. There are two for redundancy, and I do some error checking in the controller to QC the measurements prior to deciding whether to engage the heaters. The pic bellow shows the two sensors built up, the tubing I used (from ACE), the thermistor (make/part number are in the Sktech I posted a few posts back) and the epoxy used for potting.
I'm still chipping away at the plumbing. The return lines are mostly done. I still have a couple of custom PVC parts I want to turn on the lathe but it is just too cold for that now. The drains are plumbed from the overflows to the sump. I still need to decide how these terminate into the sump. And I still need to drill and plumb my filter sock/rubble box to the sump. But its getting there.
I finally got my DIN rail and started to layout my equipment panel. I still need to print up a few more sections of the cable track and I don't have all of my modules built yet that will go in here. For sure, I still need to complete the serial switch (multiplexer), the main controller and the 12VDC PSU. Here is a shot of the panel as it is now:
With the door shut:
Here is the backside of the panel. Since it is exposed inside the cabinet I didn't want bolts and what not sticking out. So I used T-nuts for mounting the DIN rails; nice and clean.
For other progress points, I got the temperature sensors for my controller put together. These are built from 10kOhm@25C 0.1% tolerance thermistors potted into stainless steel tubing. There are two for redundancy, and I do some error checking in the controller to QC the measurements prior to deciding whether to engage the heaters. The pic bellow shows the two sensors built up, the tubing I used (from ACE), the thermistor (make/part number are in the Sktech I posted a few posts back) and the epoxy used for potting.
kenneth wolfe
New member
to cool for school. I have a spare impeller I can pull dimensions off, from there we can sketch up some drawings.
Alright. PM me when you have an initial design and will provide contact info to send it to.
No update on the build for today, but I thought I would post up some details on how I'm fabricating my PCBs.
Like many parts of my build, I'm relying on my 3D printer to do the PCBs. I designed a simple spring loaded scribe for it that mounts using a single screw on the x-carriage (if you want to make one the stl can be downloaded from here). So it is easy to convert back and forth when I need to do a board. I thought about using a small mill in place of the scribe but couldn't come up with a scheme that was simple to implement and clean during board production.
To generate the gcode I'm using FlatCAM. It can take the gbr files that I get from Eagle and build a tool path that can be exported as gcode. It may run as is, but the originator of this idea (found on Hack-a-day) mentioned that on his machine he had to add a space before each Y in the gcode. So I have been doing that from the start and have not had any issues running the code from FlatCAM.
I have tried various things for a scribe-able etch resist. The original post said they were using permanent marker as the resist. But that did not provide very good results for me; lots of pitting even with several coats. My guess is that they were using ferric chloride as the etchant and because it is more viscus it has a harder time getting through small perforations in the maker. I use cupric chloride as an etchant and it is water thin. So it was a no go on strait maker. In the end I settled on a layer of Sharpie with two thin layers of india ink brushed over it. This gave a good resist against the etchant and allowed for pretty fine details to come through when scribing.
Here is a video of the scribe in action. This was a board coated with asphaltum as a resist, which turned out to be not well suited to this application. But it provides a pretty good idea of how this works.
<iframe width="640" height="360" src="https://www.youtube.com/embed/YuP2dfO_mKk" frameborder="0" allowfullscreen></iframe>
For finishing the boards, I use a stove top tinning method where I heat the board in a dedicated frying pan and use a tinning flux to coat the copper. I have seen a few folks using home made versions of tinning flux, liquid solder (as far as I can tell this is not readily available in the US), or solder paste. For me, the homemade version takes too much effort and solder paste is too expensive/perishable. But I found a lead free solder paste that is dirt cheap and widely available; Oatey No. 95. Its intended for tinning large copper pipes prior to soldering. It seems to work great on my boards and the flux compoenet can be cleaned up with hot water and dish soap.
Here is a video of the tinning process:
<iframe width="640" height="360" src="https://www.youtube.com/embed/FeW3mgyVmag" frameborder="0" allowfullscreen></iframe>
And last but not least, here is a mock up of a board at the various production stages. From left to right: cleaned copper, sharpie coat, two coats of india ink, fresh off the printer after being scribed, after etching in cupric chloride, after the resist is removed and post tinning.
Like many parts of my build, I'm relying on my 3D printer to do the PCBs. I designed a simple spring loaded scribe for it that mounts using a single screw on the x-carriage (if you want to make one the stl can be downloaded from here). So it is easy to convert back and forth when I need to do a board. I thought about using a small mill in place of the scribe but couldn't come up with a scheme that was simple to implement and clean during board production.
To generate the gcode I'm using FlatCAM. It can take the gbr files that I get from Eagle and build a tool path that can be exported as gcode. It may run as is, but the originator of this idea (found on Hack-a-day) mentioned that on his machine he had to add a space before each Y in the gcode. So I have been doing that from the start and have not had any issues running the code from FlatCAM.
I have tried various things for a scribe-able etch resist. The original post said they were using permanent marker as the resist. But that did not provide very good results for me; lots of pitting even with several coats. My guess is that they were using ferric chloride as the etchant and because it is more viscus it has a harder time getting through small perforations in the maker. I use cupric chloride as an etchant and it is water thin. So it was a no go on strait maker. In the end I settled on a layer of Sharpie with two thin layers of india ink brushed over it. This gave a good resist against the etchant and allowed for pretty fine details to come through when scribing.
Here is a video of the scribe in action. This was a board coated with asphaltum as a resist, which turned out to be not well suited to this application. But it provides a pretty good idea of how this works.
<iframe width="640" height="360" src="https://www.youtube.com/embed/YuP2dfO_mKk" frameborder="0" allowfullscreen></iframe>
For finishing the boards, I use a stove top tinning method where I heat the board in a dedicated frying pan and use a tinning flux to coat the copper. I have seen a few folks using home made versions of tinning flux, liquid solder (as far as I can tell this is not readily available in the US), or solder paste. For me, the homemade version takes too much effort and solder paste is too expensive/perishable. But I found a lead free solder paste that is dirt cheap and widely available; Oatey No. 95. Its intended for tinning large copper pipes prior to soldering. It seems to work great on my boards and the flux compoenet can be cleaned up with hot water and dish soap.
Here is a video of the tinning process:
<iframe width="640" height="360" src="https://www.youtube.com/embed/FeW3mgyVmag" frameborder="0" allowfullscreen></iframe>
And last but not least, here is a mock up of a board at the various production stages. From left to right: cleaned copper, sharpie coat, two coats of india ink, fresh off the printer after being scribed, after etching in cupric chloride, after the resist is removed and post tinning.
I cleaned up a couple SketchUp models and wanted to make them available.
The first is the cable track that I am using for wire management in the equipment panel. A few post back I showed it on the printer bed and an assembled shot. In the equipment panel shots a few sections can be seen on either side of the panel. The model can be downloaded from here:Printable cable track
The second is a DIN mount plate for the Jebao controller. This attaches to the controller using four of the screws that hold the controller housing together. It seems 1/2" 4-40 machine screws work well for this, though they are larger in diameter than the stock screws.
In my equipment panel you can see it with the controller on it in the upper left corner. Right now I am using the controller as is. But I suspect this will get hacked to work with my main control system before too long.
For those that want the model, it can be downloaded here:Jebao DIN mount
The first is the cable track that I am using for wire management in the equipment panel. A few post back I showed it on the printer bed and an assembled shot. In the equipment panel shots a few sections can be seen on either side of the panel. The model can be downloaded from here:Printable cable track
The second is a DIN mount plate for the Jebao controller. This attaches to the controller using four of the screws that hold the controller housing together. It seems 1/2" 4-40 machine screws work well for this, though they are larger in diameter than the stock screws.
In my equipment panel you can see it with the controller on it in the upper left corner. Right now I am using the controller as is. But I suspect this will get hacked to work with my main control system before too long.
For those that want the model, it can be downloaded here:Jebao DIN mount
geekengineer
Member
very cool! I like to setup a scribbling setup like that when I would get to it who knows lol.
A few updates from this weekend. I have been working on finishing up plumbing the overflows and what not back to the sump, and getting some of the cable management infrastructure installed in the stand.
For cable management I modeled up some simple hangers to give me something to support runs of wire across the sump side of the stand (the STL can be downloaded from here). They are sized large enough that they should be able to hold a number of cables and the fat end on the hook will hopefully help hold the cables in place. I threw a hole through the end of the hook just incase I need a good spot for a zip tie to deal with any stubborn cables down the road.
Here is a set of hangers fresh off the printer. I did ten, but only ended up using seven to cross the sump side of the stand. You can see them installed in one of the shots further down this post.
On the enclosed side of the stand adjacent to where the equipment panel is I mounted a section of wire shelf to hold excess cables and power supplies up and out of the way (thanks to karimwassef's post the other week for the idea). The shelf is mounted in the back with a couple screws and fender washers. In the front, I didn't have a simple hardware solution ready to go. So I modeled up a mount that slides over the two thicker lateral rods on the shelf and then has holes for screws (the STL for it can be downloaded from here).
On the sump side of things I have just about got my filter sock/rubble chamber done. I posted a few pics of the glass going together awhile back. Now I have gotten around to getting it "drilled" and in place. All that is left is to make the corian plate for the filter socks.
Here is the box in place and empty to help give an idea of how flow will go through it.
The tools used to "drill" it; Dremel, diamond bit and a spray bottle.
I'm sure lots of folks around here use this method for "drilling" tanks. But for those that don't, here is short video showing it in process. I prefer this method over using a diamond hole saw, as its easier to keep the water where is needs to be (between capillary action and surface tension it stays right in the slot being cut, note that in the video the water filled slot goes right through the glass) and there is less risk of cracking things when finishing up the hole.
<iframe width="640" height="360" src="https://www.youtube.com/embed/vYYNMRP0Kn8" frameborder="0" allowfullscreen></iframe>
Here is a shot of the sump as it is now with all the overflow piping running into it. The two that come through the filter sock/rubble box are the primary overflow lines. The two in the back center are intended to handle the surge when it goes. The final one that ends with a 90 degree elbow at the bottom is the overflow back from the surge reservoir.
For cable management I modeled up some simple hangers to give me something to support runs of wire across the sump side of the stand (the STL can be downloaded from here). They are sized large enough that they should be able to hold a number of cables and the fat end on the hook will hopefully help hold the cables in place. I threw a hole through the end of the hook just incase I need a good spot for a zip tie to deal with any stubborn cables down the road.
Here is a set of hangers fresh off the printer. I did ten, but only ended up using seven to cross the sump side of the stand. You can see them installed in one of the shots further down this post.
On the enclosed side of the stand adjacent to where the equipment panel is I mounted a section of wire shelf to hold excess cables and power supplies up and out of the way (thanks to karimwassef's post the other week for the idea). The shelf is mounted in the back with a couple screws and fender washers. In the front, I didn't have a simple hardware solution ready to go. So I modeled up a mount that slides over the two thicker lateral rods on the shelf and then has holes for screws (the STL for it can be downloaded from here).
On the sump side of things I have just about got my filter sock/rubble chamber done. I posted a few pics of the glass going together awhile back. Now I have gotten around to getting it "drilled" and in place. All that is left is to make the corian plate for the filter socks.
Here is the box in place and empty to help give an idea of how flow will go through it.
The tools used to "drill" it; Dremel, diamond bit and a spray bottle.
I'm sure lots of folks around here use this method for "drilling" tanks. But for those that don't, here is short video showing it in process. I prefer this method over using a diamond hole saw, as its easier to keep the water where is needs to be (between capillary action and surface tension it stays right in the slot being cut, note that in the video the water filled slot goes right through the glass) and there is less risk of cracking things when finishing up the hole.
<iframe width="640" height="360" src="https://www.youtube.com/embed/vYYNMRP0Kn8" frameborder="0" allowfullscreen></iframe>
Here is a shot of the sump as it is now with all the overflow piping running into it. The two that come through the filter sock/rubble box are the primary overflow lines. The two in the back center are intended to handle the surge when it goes. The final one that ends with a 90 degree elbow at the bottom is the overflow back from the surge reservoir.
So not exactly reef related, but tank cleaning for these guys and gals was why I didn't make any progress on the build tonight. Plus I thought you guys might be interested to see the other tank that resides in the tank room.
The original trio of these little guys were the by product of an inter-species tadpole competition study that was going on as I finished up graduate school. I believe the froglets came to me somewhere around nine years ago. Right now there are four in the group. Here is one of them: Dendrobates imitator (though I think there has been some reclassification and that name may have changed).
Here is a shot of part of the tank. Its a 55 gallon. If you look close there is one visible in the shot, though it is hard to see. They are quite small; two would fit on my thumbnail.
The original trio of these little guys were the by product of an inter-species tadpole competition study that was going on as I finished up graduate school. I believe the froglets came to me somewhere around nine years ago. Right now there are four in the group. Here is one of them: Dendrobates imitator (though I think there has been some reclassification and that name may have changed).
Here is a shot of part of the tank. Its a 55 gallon. If you look close there is one visible in the shot, though it is hard to see. They are quite small; two would fit on my thumbnail.
Updates and progress have been slow to come. I spent tank time this weekend working on parts for my surge valves; assembling and vapor polishing printed parts. No photos of these for now, as I would like to post the complete valve build at once. So we will just have to wait on that.
The other thing I did over the weekend was clean up a few bugs in my light controller code and validate the performance. I have another thread with the build and files for the controller (here's the link). All links in it are the most current versions, including the code. But I'll throw my test data up here as it seems as good of place as any.
I took a PAR sensor and datalogger and got some measurements from my setup. Treat these as qualitative; they were not made with any rhyme or reason to the sensor placement and there was no controlling for ambient light in the room. I hope to find some time in the near future to measure spectral distribution and light uniformity from the fixtures. But it hasn't happened yet.
In any case here is a diel curve of PPFD from running the lights on Sunday. The x-axis is in minutes from start of logging as I neglected to check to see that the logger clock was set to local time before starting measurements (rookie mistake). In any case it paints a pretty clear picture of how the dimming performs.
There are two step changes in intensity at the start and end of the curve. At the start, these correspond to the blue LEDs coming on and then the white. At the end its the reverse. These appear to represent the minimum on intensity using the default MarsAqua drivers; using my dimming circuit and the default pot-dimmer I get the same minimum. So no room for improvements there without putting different drivers in the fixture and that is out of scope for now.
Here is the test setup for those that are interest (I didn't hold the logger all day, just for the pic). You can see the sensor in the background sitting on the rock work. Its roughly in the middle of the tank vertically and off center of the LED fixture.
The other thing I did over the weekend was clean up a few bugs in my light controller code and validate the performance. I have another thread with the build and files for the controller (here's the link). All links in it are the most current versions, including the code. But I'll throw my test data up here as it seems as good of place as any.
I took a PAR sensor and datalogger and got some measurements from my setup. Treat these as qualitative; they were not made with any rhyme or reason to the sensor placement and there was no controlling for ambient light in the room. I hope to find some time in the near future to measure spectral distribution and light uniformity from the fixtures. But it hasn't happened yet.
In any case here is a diel curve of PPFD from running the lights on Sunday. The x-axis is in minutes from start of logging as I neglected to check to see that the logger clock was set to local time before starting measurements (rookie mistake). In any case it paints a pretty clear picture of how the dimming performs.
There are two step changes in intensity at the start and end of the curve. At the start, these correspond to the blue LEDs coming on and then the white. At the end its the reverse. These appear to represent the minimum on intensity using the default MarsAqua drivers; using my dimming circuit and the default pot-dimmer I get the same minimum. So no room for improvements there without putting different drivers in the fixture and that is out of scope for now.
Here is the test setup for those that are interest (I didn't hold the logger all day, just for the pic). You can see the sensor in the background sitting on the rock work. Its roughly in the middle of the tank vertically and off center of the LED fixture.
Thanks for the kind words folks. I am glad that my tinkering around is appreciated.
I have a question for those watching this thread though; I have not picked my heaters yet and am looking at options. Seems there is a wide range in prices for what on the surface appear to be comparable devices. So I'm curious what other's experiences are; Is there a brand or model that is known to be reliable? One that is known to be particularly prone to failure?
Data or documentation to back up any claims would be appreciated. But I know that is a tall order for consumer items like this (otherwise I probably would have already picked something).
Keep in mind that I am using a standalone controller and looking to run a pair of relatively undersized heaters.
I have a question for those watching this thread though; I have not picked my heaters yet and am looking at options. Seems there is a wide range in prices for what on the surface appear to be comparable devices. So I'm curious what other's experiences are; Is there a brand or model that is known to be reliable? One that is known to be particularly prone to failure?
Data or documentation to back up any claims would be appreciated. But I know that is a tall order for consumer items like this (otherwise I probably would have already picked something).
Keep in mind that I am using a standalone controller and looking to run a pair of relatively undersized heaters.
LukeGI
New member
For my 180, I purchased the Catalina 800w titanium heater with controller as it was recommended by a local aquarium maintenance/builder company that uses it regularly in their systems. The controller isn't perfect (off by a degree or two regularly from the temp monitor), however the heater itself seems to do a great job.
It can be purchased without the controller from amazon at a fairly good price.
http://www.amazon.com/Catalina-Watt-Titanium-Aquarium-Heater/dp/B004QIRX3E
Thanks for the heater recommendation. I have been looking into these and they do seem quite nice. I'm not sure I will go with them at this point though as I would like some inherent redundancy in the system; They rely on an outside controller and while that is no problem in and to itself as I am using one, I like the idea of an embedded thermostat in the heater as a secondary fail safe should my controller have some issue. The idea of a titanium case over glass is supper attractive though. Not sure what I will do yet.
For up dates there was not much progress on the actual build this past weekend. I was traveling the week before and am out again this week. So my opportunity to physically work on hardware has been limited.
I did manage to print a final version of one of the larger parts of the surge valve (its a 9 hour print) and a few sub-sections of another part to validate the geometry is correct. I had to rework a few small details and add a bit more rigidity to the motor mount. But I think all is good to go know and I will be ready to start printing all the parts in full force this weekend.
I also got my emergency lines in place. Both the sump and surge tank sit in water tight "œpans" with a drain in them. These are hard plumbed now together and through the wall to the unfinished part of the basement where the floor drain is. Its 1/2" thin wall PVC so it won't move a ton of water, but it should handle an ATO malfunction or any spills without a problem.
I have been waiting on a Mouser order that should be delivered today. So my next updates should be pretty exciting. If all goes well I'll have a functional eight port serial switch to share and maybe some details on the start of my ATO design. I am thinking at the moment that I want to build a multi-level water level sensor based on Hall-effect sensors and I have some parts coming to validate wether this will work as desired.
There are few outstanding issues that I need to address in the near future to get the system cycling but I don't think I'm that far from putting water in the system. Hopefully I will find some time this week to select some of the remaining components and get them ordered.
