Thank You, Matt, for the link to the fabrication report.

It's an interesting read, but not amusing, instead a sad literature! In fact it made me nearly break out in tears

:((
I could now start a very very long listing of severe mistakes made in design and manufacturing, but thats inutile. Everyone with a bit of knowledge about mechanical engineering can read himself how incompetent the Lifetrac fabricators are.

Its not a bit astonishing that Lifetracs don't do their job. And its a fundamental problem that can not be solved with a prototype #4 or #5. First necessity will be to exchange the complete design team!

IMHO noone involved in the design of this Lifetrac should ever design other machines. People should draw adequate consequences when they are totally incompetent for a job

Mike

Hello Matt,
You've asked for constructive critics. But thats difficult and invain if the folks building a machine do nearly everything wrong. The current Lifetrac design is poor, its now probably common knowledge.

But my last critics were focused on the manufacturing procedures used. They are just bad. You could use the blueprints of the best JD or CaseIH tractor ever built, but the result would be crap with these manufacturing standards. Its a lot of arrogance of telling people, open source machines are superior and at the same time using this fabrication practices and even publish them.
So the only constructive critics on manufacturing can be for a nearly complete change in doing metalworks at FeF. The people involved need to get a feeling for precision and learn the not-to-dos in metalworking. Open source machines don't work well without precision just because they are open source! Its simple annoying to read again and again that open source machines could be kept simple and without precision, just a stupid vision of some folks in the OSE community !

I try to give You just two examples of the poor manufacturing standards with Lifetrac. Just two from a heap of mistakes.

Example 1: wheel hub and assembly
Cutting the 2" bore in the wheel hub with a hand torch is completely inadequate. The tolerances in diameter with a hand torch are very high and the bore won't be in the center of the hub. You NEED to use a lathe for this job, hand torch cutting is insane.
Next point, the 3/4" crosswise hole to connect wheel hub and axle. Drilling this hole is forbidden, because You will end up with several tens of a millimeter play between bolt and axle/wheel hub. Hmm, ever heard about the existence of something called a reamer at FeF ???
Last example in this assembly, welding the axle to the splined shaft. Absolute no-go ! Most splined shafts are steel grades that shouldn't be welded. Noone welds something on a splined shaft. Also the wheel axle won't be straight after this welding! Remember, wheels work best when they run round and not eccentric.

I wonder how You would react if Your car manufacturer had manufactured the wheel hubs this way, providing are really bumpy ride in your car every mile, Matt ?

Example 2: cylinder mounts
Inserting a 1" bolt in a drilled 1" hole usually gives a very loose fit, so it will wear out soon under high and changing loads. Again, a reamer or a boring head have to be used.
And what about lubrification of a sliding surface under high pressure? Anyway, pairing unhardened steel with unhardened steel part will result in a high rate of wear. Is this the "lifetime design", OSE promises in its core values ?

Its not only the design, but also the fabrication that makes for a well working machine. Unfortunately many people think, poor over-simplified designs manufactured coarse without precision with stone-age-tools are great, just because they are open source.

Sorry, but I don't share that stupid belief.

Mike

Not to beat a dead horse, but you are clear on the fact that the current design is a prototype, right? As in, "a first or primitive form."

"You've asked for constructive critics. But thats difficult and invain if the folks building a machine do nearly everything wrong."

> Actually, wouldn't that make constructive cricticism easier? It seems like the only difficult decision would be where to start pointing out specific flaws, since there are so many of them.

"Its a lot of arrogance of telling people, open source machines are superior and at the same time using this fabrication practices and even publish them."

> I don't know where you're getting the claim of superiority from. I know that I personally have gone out of my way to classify the current tractor as "barely adequate."

"Its simple annoying to read again and again that open source machines could be kept simple and without precision"

> I think it's more of a design goal than a requirement. As simple as possible but no simpler. The way to achieve it is to start out as simple as possible and then add complexity only when it is required to reach performance specifications.

"Cutting the 2" bore in the wheel hub with a hand torch is completely inadequate. The tolerances in diameter with a hand torch are very high and the bore won't be in the center of the hub."

> I compiled the instructions based on information I gathered from the people who built the tractor. My understanding of the wheel hub was that the 1 7/8" shaft never touches the flat disc that the wheel bolts to. The hole in the middle is just to allow the shaft to pass through. It is the location of the 6" tube which determines whether or not the axle is centered in the wheel, and the tube welds to the face of the disc rather than passing through it. So the accuracy of the hole in the middle of the disc is irrelevant. Now that you bring up the issue of centerdness, tho, the instructions do need to include a procedure for ensuring the 6" tube and the wheel bolt holes are centered. It would be nice to do it with a CNC but a manual method shouldn't be too difficult.

"Most splined shafts are steel grades that shouldn't be welded. Noone welds something on a splined shaft."

> What problem does that create? Is there a danger of the weld failing and allowing the splined shaft to rotate? Will the splined shaft snap? What if you brazed the splined shaft in place?

"Also the wheel axle won't be straight after this welding"

> Well, all manufacturing processes are inherently flawed. It's a matter of getting ENOUGH precision, not perfect precision. Given the current design, and how the separate locations of the two axle bearings and the motor allows the axle shaft to flex and wiggle, imperfections in the mating of the axle shaft and the splined shaft seem unimportant.

"Inserting a 1" bolt in a drilled 1" hole usually gives a very loose fit, so it will wear out soon under high and changing loads."

> If it was spinning, sure, but the cylinder pins rotate through only a few degrees. Yeah, they'll wear out eventually, but not before much more expensive pieces of the tractor and replacing them is nearly free. All mechanical designs involve one or two sacrificial parts. The key is to make the consumable parts cheap and easy to replace, which the pins are.

"Is this the "lifetime design", OSE promises in its core values ?"

> My interpretation of OSE's principles is that each machine will provide a lifetime of service, not that each piece of metal will resist wear, deformation, and corrossion for a lifetime. The tractor is being designed such that it is easy/cheap to build and easy/cheap to repair and easy/cheap to upgrade. It is NOT being designed so that once built it will survive a lifetime of use without any need to repair or replace anything.

Hi Sensor,

excuse me for some emotions in that posting.
Usually I try to keep emotions out of technical discussions, but the careless manufacturing made me really sad and angry.

I know several people without professional background in metalworking and what a fantastic job they do in model building: model railroaders building live steam locomotives or RC model excavators in excellent quality. In their spare time in a garage. The gap between this and FeF built machines is more than a bit too large.

Regarding 24/7/365 use, even good commercial tractors won't stand 8700 hours for more than one or two years. But the goal i Lifetrac development was to develop an agricultural tractor for farmers, and commercial farmers use their tractors hundreds or even one thousand hours a year. Its not intended as a lawn tractor for occasionally mowing some dozen square meters. So Lifetrac should have become a tractor for commercial farmers but the design seems to be bad. Together with poor manufacturing the result of several years Lifetrac development are just crap.

that if you don't have what you like that you are using what you have

Sorry for double posting, but the forums software has cut off one half of my post

Hi Sensor,

excuse me for some emotions in that posting.
Usually I try to keep emotions out of technical discussions, but the careless manufacturing made me really sad and angry.

I know several people without professional background in metalworking and what a fantastic job they do in model building: model railroaders building live steam locomotives or RC model excavators in excellent quality. The gap between this and FeF built machines is more than a bit too large.

Regarding 24/7/365 use, even good commercial tractors won't stand 8700 hours for more than one or two years. But the goal i Lifetrac development was to develop an agricultural tractor for farmers, and commercial farmers use their tractors hundreds or even one thousand hours a year. Its not intended as a lawn tractor for occasionally mowing some dozen square meters. So Lifetrac should have become a tractor for commercial farmers but the design seems to be bad. Together with poor manufacturing the result of several years Lifetrac development are just crap.

that if you don't have what you like that you are using what you have

The biggest problem open source has is that it has yet to demonstrate that peole can may their mortgage with it. There are one or two examples of companies that primarily or entirely generate revenue from open source anything (let alone hardware) but profitability is an issue. So...it's hard to get professional-quality work done. The vast majority of open source projects are hobbies or, at best, volunteer efforts. People who know what they're doing, and who's time is valuable because of that fact, don't tend to spend their meager free time following orders on someone else's project.

People who are willing to contribute grunt work to a project they don't control, for no tangible reward, usually produce work of low quality. People who are capable of, and willing to, produce high quality work, are not satisfid unless they're in charge of the project, or at least their part of it.

So it should be no surprise that team-based open source technology projects start out at a "barely adequate" level of performance. Tons of stuff is missing, what has been done hasn't been done very well, and every solution creates even more problems. I don't see how it could possibly be any other way.

On the other hand, if you're the sort of person who's intrinsically motivated to do good things, open source tech projects are wonderful. Everything you do directly contributes, in a tangible way, to a better future. If intrinsic motivation isn't your thing, then this kind of project looks stupid.