Is it the "one man, one engine" philosophy of AMG? Or can it involve an assembly line/stations where the technicians hand-check all the tolerances, position the piston-ring gaps, etc? How much automation should be allowed in a hand-built engine, e.g., simultaneous torquing of mains or heads? Does anyone know which OEM has the most automated engine-assembly line? I'm curious which engine is touched the least by human hands.
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations TugboatEng on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Definition of OEM "hand-built" engine?
- Thread starter RETHomer
- Start date
GregLocock
Automotive
A handbuilt engine is obviously one that is filed from solid iron by one man, using files he has made from steel made from lumps of coal and ore that he dug up with a stick. Anything else is marketing.
In a modern engine plant in a first world country you use people where it isn't efficient to use robots and the like, I'd guess the various GM mega plants would come close to the highest ratio of engines per direct man hour.
Cheers
Greg Locock
New here? Try reading these, they might help FAQ731-376
In a modern engine plant in a first world country you use people where it isn't efficient to use robots and the like, I'd guess the various GM mega plants would come close to the highest ratio of engines per direct man hour.
Cheers
Greg Locock
New here? Try reading these, they might help FAQ731-376
Greg is probably right w/r/t who has the least "by hand" approach. Other candidates that come to mind would be Toyota, Nissan, and Honda, although I don't have firsthand knowledge of their production practices.
I've mostly worked in plants that are nearer the other end of the spectrum.
- Very large engines (say 200L total displacement and up) are often stall-built, where the parts are brought to the engine and a single person or a few people labor in the stall to put it all together. They'll use lifting tools, etc., because the parts are very big, but a person is making decisions about very many things in the course of assembly. There is never anyone checking tolerances on the assembly line - excluding extreme circumstances, all of the parts are within spec when they arrive on the line. Simultaneous torqing of bolts is very common, but a person is still positioning and commanding the tool. Ring gap positions are of negligible significance, but the rings have to be put on right-side-up. Subassembly of certain systems is done on separate assembly lines, but again in a very manual process (too expensive to buy robots for lower volume).
- Smaller engines, say 7-150L, commonly move down assembly lines where there are varying degrees of automation. I haven't the data to do the graph, but my suspicion is that within this size range the level of automation is largely a function of the production volume. A particular engine that I'm familiar with has piston/ring/liner subassembly done manually on a side line, while the heads are assembled from bare machined castings to the completed product without a single touch by a human. The heads are manually positioned on the engine and bolt torquing is simultaneous w/hand-guided tool.
When you ask "how much automation should be allowed in a hand-built engine" you seem to be implying that being "hand built" is desirable for some reason - I would suggest that the opposite is true from a quality standpoint, assuming the economics were a wash. In the end, as I mentioned above, economics dictate what will be "hand built" vs. "automated production" and the lower volume products suffer the quality penalty associated with hand building because it's still cheaper to fix a few of 'em than to buy robots to build them all correctly.
I've mostly worked in plants that are nearer the other end of the spectrum.
- Very large engines (say 200L total displacement and up) are often stall-built, where the parts are brought to the engine and a single person or a few people labor in the stall to put it all together. They'll use lifting tools, etc., because the parts are very big, but a person is making decisions about very many things in the course of assembly. There is never anyone checking tolerances on the assembly line - excluding extreme circumstances, all of the parts are within spec when they arrive on the line. Simultaneous torqing of bolts is very common, but a person is still positioning and commanding the tool. Ring gap positions are of negligible significance, but the rings have to be put on right-side-up. Subassembly of certain systems is done on separate assembly lines, but again in a very manual process (too expensive to buy robots for lower volume).
- Smaller engines, say 7-150L, commonly move down assembly lines where there are varying degrees of automation. I haven't the data to do the graph, but my suspicion is that within this size range the level of automation is largely a function of the production volume. A particular engine that I'm familiar with has piston/ring/liner subassembly done manually on a side line, while the heads are assembled from bare machined castings to the completed product without a single touch by a human. The heads are manually positioned on the engine and bolt torquing is simultaneous w/hand-guided tool.
When you ask "how much automation should be allowed in a hand-built engine" you seem to be implying that being "hand built" is desirable for some reason - I would suggest that the opposite is true from a quality standpoint, assuming the economics were a wash. In the end, as I mentioned above, economics dictate what will be "hand built" vs. "automated production" and the lower volume products suffer the quality penalty associated with hand building because it's still cheaper to fix a few of 'em than to buy robots to build them all correctly.
When In Lamborghini I saw 4-5 guys on a small line assembling the engines. This included dropping in the cranks/pistons/and cams.
Bmw used to/does the same with the high power stuff.
To me, the main reason for hand assembly is units per year, and if it would pay to set up a line, or just pay a guy instead.
As to which is better, it depends...Id trust any guy that puts his name on the top cover when he is finished.
F1 engines are also hand assembled, but the guys are_not ordinary fitters/assemblers, again, small numbers.
Pagani buys his engines off Merc, and proudly told me that he was the first guy they created an engine number for, Independent to all their own/other engines. These come to him nearly fully dressed, again, hand assembled at AMG. The assemblers name badge is omitted though as the intakes are different, valley on show when fitted.
I dont really see a problem with a handbuilt engine vs automation(provided Im not paying). Handbuilt also gives the guy a chance to spot any flaws in the parts or castings if anything gets missed in the quality and testing department prior to him or the line receiving the parts. The operator in charge of his engine REALLY looks at the parts well before assembling them.
On a side note ignoring numbers, Im sure if automation was better F1 wouldn't blink an eye at building a line just to assemble a few engines a year as cost is not a problem. I feel they like having as many pairs of eyes on each process as they can though.
They also crane in some of the v12 cranks in exotic stuff as they are over the allowed weight for the operator to lift.
The Veyron w16 engine is built by about four people at Volkswagen's Salzgitter engine plant, with what I would call hand assembly.
I guess it all depends who has the spanners and hammers as to which way is better.(And who is paying too!)
Brian,
Bmw used to/does the same with the high power stuff.
To me, the main reason for hand assembly is units per year, and if it would pay to set up a line, or just pay a guy instead.
As to which is better, it depends...Id trust any guy that puts his name on the top cover when he is finished.
F1 engines are also hand assembled, but the guys are_not ordinary fitters/assemblers, again, small numbers.
Pagani buys his engines off Merc, and proudly told me that he was the first guy they created an engine number for, Independent to all their own/other engines. These come to him nearly fully dressed, again, hand assembled at AMG. The assemblers name badge is omitted though as the intakes are different, valley on show when fitted.
I dont really see a problem with a handbuilt engine vs automation(provided Im not paying). Handbuilt also gives the guy a chance to spot any flaws in the parts or castings if anything gets missed in the quality and testing department prior to him or the line receiving the parts. The operator in charge of his engine REALLY looks at the parts well before assembling them.
On a side note ignoring numbers, Im sure if automation was better F1 wouldn't blink an eye at building a line just to assemble a few engines a year as cost is not a problem. I feel they like having as many pairs of eyes on each process as they can though.
They also crane in some of the v12 cranks in exotic stuff as they are over the allowed weight for the operator to lift.
The Veyron w16 engine is built by about four people at Volkswagen's Salzgitter engine plant, with what I would call hand assembly.
I guess it all depends who has the spanners and hammers as to which way is better.(And who is paying too!)
Brian,
SomptingGuy
Automotive
Another data point...
We have an assembly line here, which builds exotic engines in relatively low volumes (50/week or so). The engines are hand-built, but the assembly line is designed so that it doesn't take a massively skilled workforce to operate it.
- Steve
We have an assembly line here, which builds exotic engines in relatively low volumes (50/week or so). The engines are hand-built, but the assembly line is designed so that it doesn't take a massively skilled workforce to operate it.
- Steve
I would agree with the idea that in general, automated assembly is 'better' than human assembly. Toyota, for example, machines many of their engine parts with tolerances of <5 microns. They also have bolt runner machines that fault out with tolerances less than .1 N-m. Most experienced machinists can't even measure tolerances this tight, let alone assembly line workers.
In some assembly lines that I have witnessed, there are on the order of 10-20 'manual' processes performed from castings to assembled long-block engines. This does not include workers who operate or maintain the machines. Many of these 'manual' processes would also involve a jig or hand-held machine to perform the operation. Keep in mind that production assembly lines typically have a process time of 30-90 seconds per operation.
In some assembly lines that I have witnessed, there are on the order of 10-20 'manual' processes performed from castings to assembled long-block engines. This does not include workers who operate or maintain the machines. Many of these 'manual' processes would also involve a jig or hand-held machine to perform the operation. Keep in mind that production assembly lines typically have a process time of 30-90 seconds per operation.
The more you can factor out the human element in any repetitive operation the better off the product is. CNC for example.
Even simple things like setting work offsets and tools, is much better when automated. Much less chance for crashes and errors.
Automated engine and other components assembly, I feel will have much better consistancy and less chance of errors. Like bolts not tightened etc.
Even simple things like setting work offsets and tools, is much better when automated. Much less chance for crashes and errors.
Automated engine and other components assembly, I feel will have much better consistancy and less chance of errors. Like bolts not tightened etc.
NormPeterson
Structural
But can anybody's automation do things like work with file-fit rings and set specific spring installed heights, or is it limited to using components that have more margin against creating bad situations?
Norm
Norm
Then you get to hand rebuilt, "blueprinted" used to be the term. I took an 85hp Triumph 2L I6, and put it around 180 hp. Competition department pamphlets were a guide. I used an ultrasonic thickness tester, to allow about 1/8" milled off the head. Chambers were reshaped and cc'd to match the flat piston height in the deck. I balanced the pistons with a triple beam balance, an old master shop guy balanced the rods for total weight and mass moment of inertia, and also balance the crank, clutch, and flywheel individually, and as a match marked assembly. I spent weeks porting the head and the intake manifold for trip Weber DCOE's that I spent a month's pay to rejet and rechoke. No was the full length porting could be done by CNC.
The result was amazing, as the briggs & stratton like engine turned into a turbine smooth monster.
The result was amazing, as the briggs & stratton like engine turned into a turbine smooth monster.
patprimmer
New member
More eggs to suck guys.
Kevin
Have you looked at the top left corner of the page where it says
"INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS"
and
"Smart questions
Smart answers
Smart people"
This is the engines and fuel forum. Most here work developing engines for a living.
Oh, and by the way, what you did to your Triumph is not blueprinting, it is hotrodding.
Blueprinting ensures every part is at the optimum point within the factory specification and is used where you must race with engines to factory spec. For a hot rod, true blueprinting is an unnecessary restriction and cost.
When factory specs where all over the place it did yeild good results. I saw around 20% gains at times. Even more when loopholes where deliberately included or with little cheats like throwing in a broken drive belt for the balance shaft. It was legal for it to break, but not for you to leave it off.
With a modern precision built engine, it is quite difficult to find any power by blueprinting the mechanical parts unless the factory deliberately builds in a loophole for you or you have some little cheats.
Most tuning is now done in the engine management system.
Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &
for site rules
Kevin
Have you looked at the top left corner of the page where it says
"INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS"
and
"Smart questions
Smart answers
Smart people"
This is the engines and fuel forum. Most here work developing engines for a living.
Oh, and by the way, what you did to your Triumph is not blueprinting, it is hotrodding.
Blueprinting ensures every part is at the optimum point within the factory specification and is used where you must race with engines to factory spec. For a hot rod, true blueprinting is an unnecessary restriction and cost.
When factory specs where all over the place it did yeild good results. I saw around 20% gains at times. Even more when loopholes where deliberately included or with little cheats like throwing in a broken drive belt for the balance shaft. It was legal for it to break, but not for you to leave it off.
With a modern precision built engine, it is quite difficult to find any power by blueprinting the mechanical parts unless the factory deliberately builds in a loophole for you or you have some little cheats.
Most tuning is now done in the engine management system.
Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &
for site rules
- Status
Similar threads
- Locked
- Question