Substantive Final report - CDP Phase 1
The substantive final report provides evidence that you have met the documentation requirements, we expect this to be largely composed of links to documentation or forum posts with some contextual information.
Objectives
What we set out to do.
To quote our application to the CDP:
Our goal is to make a system that is:
- well documented,
- easy to make, use and hack,
- extremely affordable,
- and versatile (protocol-wise).
To these ends, we will:
- (1,2) make multimedia documentation covering most aspects of the project,
- (2,3) continue relying on widely available components, 3D printing, and modularization;
- (3) use cheaper parts and make use of existing equipment (i.e. the micropipettes),
- and (4) modularize hardware and software, implement a toolchanging system, and interface with other OScH products (thermocyclers, turbidimeters, etc.).
Results summary
What we did.
-
well documented
: we have documented part of the assembly process, and most (if not all) of the development. Lots of pictures were taken here, most of which will be used to finish the assembly guide. A part of the assembly is still undefined, and will only be documented fully when Pipe's enclosure is designed and built. -
easy to make, use and hack
: though we are still missing a full assembly guide, we have kept our project as simple as possible, and still choose widely available components. Some simple soldering is required, but hopefully not for long (see #47 (closed)). -
extremely affordable
I don't believe there can be a pipetting robot half as cheap as ours :) -
versatile
: the tool-changing system has been implemented, and the hardware is relatively modular. We have not had time to write interfaces with other OScH devices, but it is still on the roadmap.
Posts on the GOSH forum:
- First post (after the grant was awarded): https://forum.openhardware.science/t/pipetting-bot-gosh-coldev-round-1-presentation/3797
- Project updates: https://forum.openhardware.science/t/pipetting-bot-gosh-coldev-round-1-updates/3951/
Collaboration summaries
Collaboration logs have been kept in a few issues. A substantive summary can be found at the last comment of each one.
- Solomon: https://gitlab.com/pipettin-bot/pipettin-grbl/-/issues/20#note_1095470643
- Pipe: https://gitlab.com/pipettin-bot/pipettin-grbl/-/issues/19#note_1095497872
- Nico: https://gitlab.com/pipettin-bot/pipettin-grbl/-/issues/28#note_1095503467
- Renan: https://gitlab.com/pipettin-bot/pipettin-grbl/-/issues/18
- Related issues tag: gosh-col-dev.
Documentation
We have produced the following documentation:
- Quite thorough developer's documentation: https://gitlab.com/pipettin-bot/pipettin-grbl
- Organized and hyper-linked README.md files, per directory.
- These files describe each aspect of the project individually, and some of the interactions between them.
- Unfinished assembly instructions:
- Source files: https://gitlab.com/pipettin-bot/pipettin-grbl-docs
- Published content (work in progress): https://pipettin-bot.gitlab.io/pipettin-grbl-docs/
- Photo album: here.
Design files and software
- Published at the main repository: https://gitlab.com/pipettin-bot/pipettin-grbl
Discussion and next steps
Several aspects of the project took much more time than expected:
- Managing the grant: making sure that suppliers complied with the administrative requirements of our fiscal sponsor was not trivial. We could not purchase the parts directly (this also took time and was error-prone) and no supplier had all parts in stock.
- Delays: (1) Finding a suitable fiscal sponsor was difficult. (2) I misinterpreted the CDP's terms regarding eligibility of collaborators, and this led to a significant delay. Once it was cleared up, I immediately contacted Renan, because I knew that he could be trusted with the collaboration. (3) However he had to leave the project for personal reasons a few weeks later. Later on, (4) I realized that the remaining collaborations were moving much slower than what I had expected. (5) Finally, our 3D printer started malfunctioning during the last few weeks, which delayed testing Pipe's tool-changer.
- Writing updates in the GOSH forum and preparing reports also took a significant amount of time.
- Complete documentation is an excellent goal, but also very time consuming. I ended up pausing the GitBuidling documentation when I realized I had to spend hours writing the part libraries (before writing the documentation proper).
Thoughts around the CDP:
- Follow-up meetings with the CDP working group would have been really nice, but none were scheduled or proposed.
- While partnering with OHM seems like a nice idea, it will probably only demand more time from the awardees.
Lessons for the future:
- Rely on pre-existing collaborations (new collaborations take time).
- Purchase all parts from a single supplier (to reduce the administrative overhead).
- Write the assembly guide in regular markdown, and move it to GitBuilding only after it is done.
Pending work from phase 1:
-
Collaborators have not yet delivered their final designs. -
The designs from collaborators will need testing. -
Volumetric calibration of the pipettes. -
Write the usage and assembly guides.
Future work:
- Move to the Klipper CNC firmware.
- Adapt OT's protocol designer to our machine.
- Design our own pipettes.
Edited by naikymen