Back to Helium Frog                           Blog                   CAD Models                     Details           Other Robots


   The Helium Frog Delta Robot you see here was the first delta robot in the world printed from reprap parts. This basic machine was the inspiration for all the other delta low cost robots that are now sold worldwide. Work began in early 2010 and by October of that year the basic machine was up and running. The delta robot concept is not new as the format was first suggested in the 1970's, however this type of robot was out of reach for the home builder. Adrian Bowyer had developed the home Cartesian robot which inspired us all, but there hadn't yet been a delta design as far as I knew, so I designed one myself on Catia V5 CAD software. My initial thoughts were to design the robot for simple pick and place jobs and first off I fitted it with a pencil for drawing. I did leave room for an extruder, but the design proved too inaccurate for any 3D printing. It should also be noted that the bowden extruder which is common today had not really been seen at this time was not available until I had completed the first assembly. This is why the frame is very strong and rigid as it was designed to take an extruder mounted on the moving head as all previous reprap machines had. This innovation proved cruical to the later machines as it reduced the mass of the moving head and made part printing much more rapid. It also enabled the frame design to be reduced in size and made more simply.

   By todays standards the Helium Frog Delta is quite primitive, slow and inaccurate, but it did at least prove out the concept and the basic software algorithms created for it enabled others to advance the design much more quickly. I had initially discussed my concepts on the Reprap Forum, but no one seemed very interested, however after I posted a video on youtube of the machine making its first moves interest started to pick up.
I had open sourced the models and software hoping that others could build on this concept.

   Johann C Rocholl was one of the first to contact me. We discussed the weaknesses and how the design could be improved. The first was of course to replace the lead screws with a belt design and improve the simplicity of the frame. It's really amazing to me how this wasn't obvious when I designed the MK1, but that is of course how new designs are, when you see one in the flesh its clear how things can be made better. I had begun these changes for a MK2 design, but seeing the rapid progress of Johanns work and the ideas he had for improving almost every part I felt comfortable that the concept had taken hold and so moved onto a SCARA robot concept which again I hadn't seen done. He also managed to use the newly designed bowden extruder from another reprap design so the machine became able to print its own parts (a feat which mine could not). This new machine was a massive leap in technology.
   Johanns "Rostock" design proved very popular and many people printed the parts to make their own. So with this the delta concept went global very quickly with all sorts of startup companies taking up the baton.

   Today you can pick up a delta robot kit for around 250. They come in all shapes and sizes and each have unique features. These modern designs have self levelling, LCD screens, heated beds etc. and are very rapid. It is a testament to Open Source Design, collaboration across the internet and the enthusiasm of individuals that from such humble (And shaky!) beginnings withing a few years a new concept can be turned into a global industry.

   If you wish to read how things were done check out the blog or download the 3D model from the links above.


This page is dedicated to the development of the Helium Frog Delta Robot. The design is mostly complete and if you would like to help with the development please feel free. Some basic .stl models are available here to get you started. I will try to post the latest designs soon. Check the blog page for the latest progress.

If you would like to see it working, check out the video on youtube by clicking the link below

I have a craving to build another delta to replace my prusa style printer. With this in mind I purchased a Sintron Technology delta kit from ebay. I suppose this could be considered a great great grandchild of the Helium Frog Delta and of course the improvements over the original ar amazing.
The robot is mostly built, but in order to keep the cost down sadly the kit is very poor in places. The diagonal arms were the first area of concern as the threaded rod did not enable the arm length to be set accurately. I purchased an aftermarket carbon fibre rod kit and this seems to have improved things.
Next is the bed levelling probe. This is very poor and uses an allen key for the probe and a very flimsy latching mechanism. I am just in the process of designing a new mechanism which I think will be unique.

Lastly the spool holder is also very poor and there is no place really to mount the electronics. I have redesigned the top of the machine to take a proper spool and all the electronics will be on the upper part of the machine. 3D printers tend to get very dirty near the print bed with small filament shavings. so the top of the machine is the natural place for the power supply, LCD and Ramps controller.

Progress on the Sintron Delta has been quite slow, but steady. The kit is really very poor, but luckily I have now built quite a few machines, so sorting it all out is getting a little easier. I have totally redesigned the bed probe. I still have a few more parts to make.
  The recommended "Rocholl" firmware seems to have been modded beyond all recognition and the latest build has  developed to such a point that it accounts for every option possible and as such has all the wrong configuration for my machine. I even had difficulty compiling it on the Adruino IDE and really didn't want to wade through correcting all the build errors. The Marlin firmware has options for everything. The Repetier firmware seems to be the same and also has a configuration setup utility and some sort of server thing going on. Is it any wonder that many beginners are confused as to what to do. I have been building machines for years, and it stumps me most of the time. This is why many people now buy a machine that just works out of the box. It's a shame as you don't learn anything this way and I am sure many skilled people who could contribute are put off doing so.
   A bit of research led me to . They have some firmware that I was able to understand (It's based on Marlin firmware but compiled without issues). I uploaded this and am most of the way to getting the printer working. The heated bed didn't work initially, but I found I needed to run two sets of wires from the power supply to the Ramps board. I now have the robot calibrated (The extruder settings seem a bit odd, but I measured the filament length and adjusted the values this way. The robot homes and moves as it should and the XYZ scaling is correct. The print head moves across horizontally and is not dished which is a common problem on deltas. There should be no reason why I cannot print next weekend. I still need to complete the bed probe as I need to make a fixture that can remove the new magnetic probe end after calibration. This should be fairly simple and is just a turned bit of plastic. I then need to modify the firmware G29 code routine to pick up the probe, measure the bed heights and then return it back to the holder. I have had a look and this seems fairly straighforward. The next task is to dial in the print settings. I am hoping I can get some gcode generation software where the settings are fairly close, but this is never usually the case. I also noticed the filament seems very runny. I havent printed PLA before and maybe this is how it is. I will try to check the actual temperatures of the head. Maybe I have the wrong thermistor type selected. Unfortunately I don't know what one was supplied in the kit.
 Once the machine is up and running I will clean out all the firmware (Remove language options, options for other control boards, options for cartesian robots.....the list goes on) this way I can fault find more easily and upload any updates in shorter time. I hope it makes good parts!