Author Topic: Mercedes-Benz W165 Grand Prix engine in 1:3 scale  (Read 57618 times)

Offline Hugh Currin

  • Full Member
  • ****
  • Posts: 473
  • Klamath Falls, Oregon, USA
    • www.currin.us
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #165 on: July 30, 2017, 06:59:56 PM »
Mike:

Sorry for the hard question. :-)

And thanks for the explanation. Using a combination of "hand coding" and 2D contours, fed through a CAM, is likely the best solution. As you say though, very labor intensive and prone to small errors. It takes me the longest time for the simplest "hand" generated code, and still likely end up with errors. I do better with a CAM program, but get nervous when cut and paste is involved. Have you tried CAMotics to test G-code programs? I've used it a little and it seems to be useful. (Linux program) I haven't found a good prototyping material that works for me. Most, including machinable wax, cost nearly as much as aluminum. I tried to make machinable wax which worked OK, but found "chips" get ground into the floor and hard to remove.

I have the same dilemma regarding 3D CAD and CAM packages. I'm also using Linux exclusively and am avoiding getting a Windows box. I haven't found a native Linux package that looks good, at any price. There are some projects out there to create one but they are in their infancy. CamBam has the start of profiling but it's rudimentary and hard to use.

I tend to use "manual" milling as much as generating a program. I use MDI commands a lot. I have trouble stopping at a given point, especially with a digital readout. MDI commands allow precise motion one command at a time. I'm still getting use to my Sherline CNC lathe but similar so far. Also like the jog a certain distance (0.10", 0.010", 0.001" etc), very nice for setting up a job.

I use VariCAD for my CAD package. It has some nice 3D functionality but not like to really high end packages. It is native to Linux and I do like it. It does have a STL output but I haven't looked into the format of those files. That may be a possible scheme for profiling.

In CamBam there is  a "Roughing Clearance" (works roughing and finishing) that leaves, or takes away, a given amount from the specified boundary. I find this easier than changing the tool diameter. So many ways to do things.

Absolutely magnificent work on the engine. Even more so now that I know you're using 2 1/2D software. I will continue to learn from you.

Thanks again.

Hugh
Hugh

Offline Vixen

  • Full Member
  • ****
  • Posts: 1180
  • Hampshire UK
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #166 on: July 30, 2017, 08:20:11 PM »
Hello Hugh,

It works two ways, I learn from you and you learn from me. You showed me the way with the Intel D525MW mini computer and the shaft encoder for the lathe. Hopefully, together we can also spread the CNC gospel to others on this wonderful MEM forum.

Actually I very rarely hand code, I tend to edit or modify the code generated by the CAM program if and when required. Mostly I use the code straight from the CAM. I use DesKAM (don't think it is still available to buy) which allows me to visualise the tool-path. The tool-path visualisation in LinuxCNC AXIS is also very good, you can view the proposed tool-path from any plane as well as in 3D. Either or both allow you to see any glaring errors and to correct them in time.

My AutoCAD and DesKAM are on a Windoze machine. I still run under Windows XP, which is regarded by many as the best and most reliable of brother Bill's offerings. Since XP Windoze has gone downhill fast. I have found nothing comparable in native Linux.

Yes, the LinuxCNC MDI command is very powerfull, especially the ability to machine up to a given point. Did you know that you can enter/copy several command lines into the MDI box. They are stacked and are worked through in sequence. For instance you can machine to a given point, lift the tool and then return to a starting point to await your next depth command. Another trick I learned is to peck-drill deep holes using the jog function. Jogging downward in five thou (0.005") breaks the chips efficiently, there are do long stringers attempting to seize the drill in the hole.
 

The trick of amending the true tool diameter is intended as a way of refining the diameter of a hole, in much the same way as making a small adjustment to a traditional boring head. Most CAM programs, even the basic low cost packages, offer a roughing clearance option. Over the years I have found ten thou (0.01") to be a nice comfortable amount of material to leave for the finishing pass. Works for aluminium, mild steel and brass.

Cheers

Mike
It is the journey that matters, not the destination

Offline Roger B

  • Global Moderator
  • Full Member
  • *****
  • Posts: 3876
  • Switzerland
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #167 on: August 01, 2017, 09:30:37 PM »
Splendid as ever  :praise2:  :praise2: The coffee cup gives a good guide to the size.

I am stuck at the manual machining level for much the same reason you are staying with 2/2.5D. At the moment I don't have the time to invest in the learning required  ::)
Best regards

Roger

Offline Vixen

  • Full Member
  • ****
  • Posts: 1180
  • Hampshire UK
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #168 on: August 01, 2017, 09:35:46 PM »
Hi Roger

Thanks.

That coffee cup also sums up my mood at the time.

Mike
« Last Edit: August 01, 2017, 10:03:24 PM by Vixen »
It is the journey that matters, not the destination

Online Art K

  • Full Member
  • ****
  • Posts: 1116
  • Madison, Wisconsin USA
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #169 on: August 02, 2017, 03:08:35 AM »
Mike,
Still following along. Looking great as well. Until Hugh chimed in I didn't realize you were not using 3D cam. Having a Tormach mill with the capability to do more than I am able makes that seems like a lot of work.  Sometimes I run it an inch high, make sure it's doing what it should (especially on a long part) then stop rewind the program drop it to Z0 & run it for real. I don't tend to use a scrap test block. I have the jog controler with the turn dial for +or- moves and this is real handy. I do tend to do a lot of manual and mdi moves. Sometimes that is just easier than setting up a program. I'm with you on the learning curve though, it's taken me a while to get this proficient with what I have and don't want to start over at ground zero with something else.
Art
"The beautiful thing about learning is that no one can take it away from you" B.B. King

Offline mikemill

  • Full Member
  • ****
  • Posts: 266
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #170 on: August 02, 2017, 10:04:29 AM »
Mike

I too am following your amazing project, Hugh was asking what material to use to test programs, I use MDF, I glue stacks of 18mm to emulate a billet, it cuts cleanly although  it get a bit furry in the middle of the sheet. Point being itís cheap and readily available, it has proved invaluable in proving the valve port angles on the Triumph cylinder head.

Mike

Offline Vixen

  • Full Member
  • ****
  • Posts: 1180
  • Hampshire UK
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #171 on: August 02, 2017, 11:23:29 AM »
Hello Art and Mike

Thanks for dropping by, it's good to have a chat.

Art, You have a Tormach mill, are they as good as everyone says they are? I would dearly like one to replace my small EMCO F1 mill, the 8" x 4" table travel can be limiting. When you add to the purchase price, the shipping costs to get it to the UK, then add a further 20% import tax on top, well, it just ain't going to happen. Do you have the Mach 3 or Pathpilot control software? Pathpilot is pure LinuxCNC but with a better and machinist optimised display screen. Underneath they are the same software.

Did I tell you that my EMCO F1 originally belonged to Her Majesty's Prison Service. It was used in a vein attempt to retrain the prison inmates for a better life. There were some very 'tough nuts' in Parkhurst Prison. When the little used machines were sold off, I was the lucky guy who got them. Since then I have replaced the steppers and drivers with more modern stuff from China and converted to LinuxCNC. I have also built a stepper rotary table and a 4th Axis unit based on a miniature lathe.

I also have a jog controller with a digital handwheel. I cannot get on with it. I much prefer to tap on the keyboard. I now have an index finger with the speed of a woodpecker.

Like you, I will normally test run new tool-path code by machining fresh air before I commit to the metal. Only once or twice have I machined a test piece and then it was to check out some complex pseudo 3D shapes with lots of hand code/editing.

Mike, I have been following your Triumph build with great interest. One problem with MDF and also to a lesser extent, my preferred Acrylic/Perspex/Plexiglass, is that these cheap substitutes can be very abrasive. I have found that MDF will wear HSS tooling faster than aluminium. Ceramic and carbide tooling is much harder and are not effected in the same way. Years ago, my previous company decided to test run a large radar dish program using a glued stack of chipboard......And you cannot find a less suitable test material than chipboard. They only got half way through the test, the chipboard was so abrasive it destroyed the cutters.

Lets go and make metal chips

Mike
« Last Edit: August 02, 2017, 12:01:59 PM by Vixen »
It is the journey that matters, not the destination

Online Art K

  • Full Member
  • ****
  • Posts: 1116
  • Madison, Wisconsin USA
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #172 on: August 02, 2017, 12:44:28 PM »
Mike,
I like the Tormach and have found it is generally capable of more than I.  :-\ I still have mach3 installed but am in the process of changing to PathPilot. I gave the specs to my local computer shop and he's gonna check into building a used computer for me.
Oh yeah, I went over to Tormach with a trailer and picked mine up.
Art
"The beautiful thing about learning is that no one can take it away from you" B.B. King

Offline Vixen

  • Full Member
  • ****
  • Posts: 1180
  • Hampshire UK
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #173 on: August 02, 2017, 01:12:32 PM »
Oh yeah, I went over to Tormach with a trailer and picked mine up.

Oh, how I wish.

Pathpilot uses LinuxCNC. LinuxCNC is distributed under an 'open source' licence agreement, so Tormach are obliged to share their Pathpilot software under the same agreement. They are NOT obliged to give details of the proprietary changes they made to the software (display screen etc) for their machines. I obtained a copy of Pathpilot from Tormach, The good people on the LinuxCNC forum helped me make some changes so it now sort of runs with my non-Tormach machines.

I like the Pathpilot operator interface, much more machinist friendly than LinuxCNC's Axis display.

The thing I discovered is that Pathpilot needs a much faster and higher spec computer than that required for LinuxCNC, That's strange since they are running basically the same core software, it must be the graphics for the new display which requires the extra computing power.

I do not know what specification your computer shop are building to. Best to be careful and check with Tormach first. Please let me know, as it may help me get my Pathpilot up and running.

Regards

Mike
It is the journey that matters, not the destination

Online Art K

  • Full Member
  • ****
  • Posts: 1116
  • Madison, Wisconsin USA
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #174 on: August 02, 2017, 02:06:18 PM »
Mike,
I have a tech sheet from Tormach with specs for what they are sure works. When I'm home from work I'll forward that to you.
Art
« Last Edit: August 02, 2017, 07:46:56 PM by Art K »
"The beautiful thing about learning is that no one can take it away from you" B.B. King

Offline Vixen

  • Full Member
  • ****
  • Posts: 1180
  • Hampshire UK
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #175 on: August 05, 2017, 05:03:07 PM »
Part 9D       Crankcase: the final machining stages

In this installment from the Vixen's Den we reach the final stages of machining the crankcase, before the long process of hand finishing those parts which could not be reached by the machine tools.

The engine's oil breather system is located on the top of the crankcase between the cylinder banks. The air from within the crankcase, laden with oil mist droplets, is expelled upwards into the plenum chambers. In the chambers the air velocity suddenly drops, causing the entrained oil droplets to condense and fall back into the bottom of the crankcase. I started by milling the internal details of the cover plates.




Here the cover is temporaly bolted to one crankshaft half prior to machining the first cylinder face




The right half of the crankcase is bolted to a 45 degree angle plate and indicated to be square. The cylinder face is faced off flat to the finished dimension, then the cylinder base flange outline was profiled. The outlet ports for the crankcase breathing system are becoming visible.




Slightly out of sequence, I know. The top of the gear-case has some detail added. The water coolant transfer passage, from one side of the engine to the other, has the pipe flange machined.




The two bumps which define the outside of the auxiliaries bevel drive are machined next. The outside shape is a series of contours spaced at 10 thou (0.01") intervals. They took a long time to hand code, so I did a trial cut on a block perspex (acrylic) to prove to myself that there were no glaring errors in the tool path. It was a nervous time when I started on the crankase, The shell thickness of the housing was only 1/16" (1.5mm) around the previously machined internal cavity. I feared that at any moment the ball cutter would break through and the whole crankcase would be ruined. My luck held for another day. Measure twice, cut once they say, I must have done that a dozen times before starting this particular cut.




There were still a few tricky bits of machining to complete before the two crankcase halves are finally joined and finished. Two oilways for the waste oil scavenge system need to be drilled down the centre of the spine which separates the crankshaft space from the oil breather plenums. The oilways are 3mm in diameter and over 4" (100mm) deep. There was a risk that the long thin drill would wander off course and break out in some unwanted location.

I mounted the crankcase vertically against a robust cast iron angle plate. I also needed to raise the mill head another 4 inches to allow for the very long drill. I used a 1/2" plate with offset holes to raise the mill head high above the mill table, there was only just enough power lead to reach the motor when the head was raised to full height. Got lucky again.

The technique I use when drilling long thin holes is to start off with a centre cutting end mill of the correct diameter and drill as deep as that mill cutter allows. This accurately positioned hole then guides the drills that follow more accurately than if I had started with a centre drill. Accurate alignment of the start and frequent clearing of the drill flutes is key to an accurate deep drilled hole. I follow the end mill with a new stub drill, then a new jobbers length drill and progress to the longest drill. I always try to have the minimun of unsuported drill between the chuck or collet and the work-piece. Minimising this distance also minimises the possibility of the drill bending in the open air and with it the risk of the long drill deviating off course. Retracting the drill every 0.1" to 0.2" is an essential part of the operation, a squirt of WD40 down the hole helps lubricate and clear the drill chips. I always 'peck drill' which breaks the chip into very short lengths, it is the long stringers which can jamb in the flutes and break a drill deep in the hole.

Well, I was carefull and got lucky again. I drilled four holes over 4" deep and estimate all were within 0.020 thou (0.5mm) of true.






Finally it was time to bolt the two crankcase halves together and finish the last remaining cylinder face. I made sets of special bolts out of high strength aluminium alloy (HE15). the bolts were made well over length with tommy bar holes to get them real tight. I used plenty of JB Weld epoxy to seal the joint faces and grout the bolts in place. The top cover received the same treatment. I gave the epoxy several days to fully cure.

The excess bolts heads were sawn off and the assembled crankcase was mounted against the vertical angle plate and the last cylinder face was faced off and the flange contoured in the same way as the first. The aluminium bolt heads were almost invisible after the cylinder face was machined.






The final machining operations were to make and fit the clutch bell housing flange to the rear of the crankcase. The rear flange plate outline shape was profiled then attached to the rear of the assembled crankcase by more aluminium bolts and JB Weld epoxy. The crankcase was again mounted against the vertical angle plate and the internal detail was milled into the flange plate. When completed, the aluminium bolts were again invisible.






That's as far as I can get with the machining. From here on it's over to the Dremel and Swiss files to smooth and blend the surfaces that could not be reached with the machine tools. It has been a long slog to get his far, about 80% of the billets I started with have been reduced to chips. The hand finishing will be a well earned rest from standing in front of the milling machine for endless hours.

Next time I will have finished with the crankcase and about to commence on the next items to be made.

Stay Tuned

Mike


It is the journey that matters, not the destination

Offline simplyloco

  • Full Member
  • ****
  • Posts: 220
  • Lots of Rebuilds in progress!
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #176 on: August 05, 2017, 05:20:07 PM »
Astonishing work. I didn't realise there was that much talent not ten miles from me! I'll be over...
John
PS. And I thought my Brit loco was nice until I saw this thread!

Online michelko

  • Full Member
  • ****
  • Posts: 190
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #177 on: August 15, 2017, 02:10:59 PM »
Holy cow
very impressive  work

Offline mikecole7

  • Full Member
  • ****
  • Posts: 34
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #178 on: August 15, 2017, 02:17:32 PM »

               Excellent work Mike. You a master with CNC.
               See you at Bristol show

            Mike   

Offline Vixen

  • Full Member
  • ****
  • Posts: 1180
  • Hampshire UK
Re: Mercedes-Benz W165 Grand Prix engine in 1:3 scale
« Reply #179 on: August 15, 2017, 02:58:00 PM »
Mike and everybody,

I'm realy looking forward to the Bristol Show this weekend. I hope to meet up with all of the usual suspects. Call in at the Internal Combustion Engine Builders stand and say hello.

Unfortunately, I don't think Dr Jo will be able to make it this year.

CNC does not make it easier, but it does allow you to consider more complex shapes.

Mike
It is the journey that matters, not the destination