Reverse Engineering Cost Alone

[beefy]

Was given a couple of brackets to reverse engineer. I'm surprised at how long it takes to get an accurate reproduction of just one piece.

Stick on paper with double sided sticky tape.
Accurately draw around perimeter, inside holes, inside slot.
Measure hole diameters.
Scan and bring into Coreldraw as a bitmap.
Position holes over bitmap.
Manual tracing of everything else.

It's taking well over half an hour to do just one. Remember accuracy is important here as this is reverse engineering.

What do you think is a fair price just for the reverse engineering on one bracket, i.e. no cutting. This bracket is about 14" x 5".

Keith.

PART A.jpg

[gamble]

The last piece I had that was about that size I took it to fedex/kinkos and had then scan and email me the image. Then I imported and traced it. Took 10 minutes but it was a much simpler design.

[beefy]

Thanks Gamble.

I don't know if I'd like to rely on a scan to maintain accuracy. I attempted a scan at home and all the edges came out blurry. In any case by the time I'd got to/from the scanning shop I'd be able to achieve the same end result at home with a pencil trace and scan, plus not pay them, and fuel costs.

I've got a camera on order for doing optical tracing but just have to use this workaround for now.

Keith.

[vmax549]

Here is my take, When ever I have a job that I do not have the proper tools to do it efficently I charge based on how long it would take IF I had the proper tools. In this case there are several ways it could be done. You have the right idea to go to optical scanning of the parts. I suspect you will use SheetCam for the scanning? It looks to be about a 1 hour job. that includes drawing the part out on the cutting table to see IF it is correct. Use a pen and draw the part then lay the part on top of the drawing.

NOW can you get blood out of your turnip(customer) only you will know that. Sometimes you have to donate to the cause and just charge what you can get out of it. Sometimes you win and sometimes you loose. Just depends on the customers needs and how bad does he need it.

(;-) TP

[beefy]

Cheers Terry,

can't argue with that one (proper tools). And yes I'll be using Sheetcam with the camera. Just have to source some black water washable paint or hair spray now, and hope my kids don't steal it from me

I've got two different brackets to do so for the time invested in each one I was thinking $30 each for the reverse engineering then about $15 per bracket to manufacture. Holes won't be cut, they'll have their centres marked by blipping a centre punch mark with the plasma torch, then I'll drill them out afterwards for accuracy and appearence.

I like that point about drawing it out with the plasma table and checking it matches the part. That helps to "sell" the reverse engineering side of things. If the part is within A4 size I can just print it out and check it, but I won't tell the customer it was that easy . My printer is quite accurate (use it to make my circuit boards).

The guy wants to chrome plate these then sell them here.

Keith.

[dustywill]

Cheers Terry,


I've got two different brackets to do so for the time invested in each one I was thinking $30 each for the reverse engineering then about $15 per bracket to manufacture.

Keith.

I think you are too low on the reverse engineering. If you are doing it accurately you will be the cheapest game in town for something like this. Not a bad position and if you are happy with $30 then have at it. It is a one time cost for your customer. $60 (about an hour) should not be too much.

As for the paint, what si ti for? If you are darkening the part for the scanning, you may consider layout dye or something like it. There is a cleaner that gets that stuff off well.

[beefy]

Thanks very much DustyWil,

I'll Google for some of that layout dye. Half the problem a lot of the time is knowing what exists and what to look for.

Keith.

[vmax549]

Water Soluable Spray "Hair Dye". It will wash off with water.

(;-) TP

[vmax549]

Water Soluable Spray "Hair Dye". It will wash off with water.

(;-) TP

[beefy]

Thanks Terry,

that's what I've been looking for, seen it mentioned on Les site. I Googled for it but didn't have any luck. I'm guessing I might be able to find it at fancy dress shops maybe.

Keith.

[vmax549]

More likely to find it at costume shops (;-)

(;-) TP

[Brand X]

Holes won't be cut, they'll have their centres marked by blipping a centre punch mark with the plasma torch, then I'll drill them out afterwards for accuracy and appearence.
Keith.

Have you thought about marking the holes this way? Plasma drill mode, but marking instead..(Little divot not punched through) This is with my Cutmaster 120 amp machine with a low-end of 30.. My Esab has 10 amps bottom, and it works sweet in it's own way.. I put a M00 pause to change out/ amps/consumables/when cutting the outline..

[beefy]

Thanks BrandX,

but I've actually been down a very LONG learning road and developed a dedicated circuit board to go a lot further than that. One of the issues with what you described was waiting for the post cut airflow to stop after a hole mark. My circuit board uses a closed loop method instead of just a barebones time delay so it commands the torch to turn on (i.e. while post cut airflow is on), monitors the cut current and when that's detected, it starts a time period. See the linked Youtube video to see several markings done in succession, no more post cut airflow headaches.

The board completely takes control of the TORCH ON signal. In addition I've added adjustable current sensing so even if my plasma cutter was set at its max of 80 amps the hole marking can be done at whatever amps you decide, 15 amps if you wanted. Additionally there is a time adjust so once your set current is reached you can adjust the time period that the torch is on. I can get my "centre punch" marks so small I have to look for them. Marking at a much lower current should help a lot to improve electrode life if doing a lot of hole marking.

With Sheetcam and code snippets and an aux output, I can create gcode to automatically put my board into "hole mark mode" and wherever there's a hole, a "centre punch" mark will be put in its place at the hole centre. I use a separate tool and have the marking height at 1mm for best accuracy.

The Youtube video is hole marking and cutting my plasma table slats. There's was also a zillion "shark tooth" toppers I cut and also marked their holes too. Drilled 6mm holes in everything (over 700 plasma marks) and put 6mm bolts in. Every one lined up so I was pretty happy. I also only used one electrode. Jim colt told me that it's the initial turn on (pierce) that erodes the hafnium in the electrode so I'm guessing my low current hole marking is helping to extend the electrode life.

https://youtu.be/yg-s-EAqMfI

Keith.

[Brand X]

With the SL-100 SV torch, it does not seem to matter on the post flow. It's fast.. I like using different consumables, (depending on machine) for the marking part.. I put the delay in to change out the amps, and or consumables.. If I was doing tons on them, then maybe a circuit would have benefit, but it's perfect for just a few I would do.. Nothing to add, and the divot is just right for my Ironworker punch..(Mag drill too)

[beefy]

Yeah, I've heard some torches work with the basic method and others don't. The post cut airflow on the Hypertherms is fixed at 10 seconds so once the torch is fired and shut off, you have to wait the ten seconds before you can send that short duration ON signal again. So just 10 hole marks added 100 seconds, which got old real quick. I used to have to put the 10 second delay between hole markings to make sure the post cut airflow had turned off.

Keith.

[BensPlasmaAu]

Hi beefy,

Is the camera going to be used with scananything from sheetcam? I'd be interested to know how you go with your setup.

Ive been thinking as well as to how to go about and replicate accurate parts from examples. For one I've been thinking about a layout board based on an engineers drafting table where you put the part in the corner and use that as its origin. It would have a sliding ruler and measurements.

Another idea to get correct dimensions of center to center distance of holes was to take a set of vernier calipers and make two pins with a 45 degree point and fit them over or instead of the jaws. The points of the 45 degree pins match up with the edge of the vernier jaw and will naturally find the center of the hole, or close enough if the hole is elongated. Then its a matter of figuring our the hole dimension, and that's the easy part.

Anyway, it makes sense to me (I've got the image in my mind and explaining it to people seems to confuse them) and I wish I had my workshop up and running to start creating this sort of stuff. Also I need more experience on CAD to get ideas on paper.

Ben