Small Production Runs

[Randy Webb]

There have been a lot of items that this board has been a sort of central information exchange on. Matt's 964 conversion hinges, the Ti exterior hinges for R type deck lids, various CF tidbits, and a lot of other things.

There are a lot of skilled and innovative people who make something for themselves, then wind up making a few, dozens, or even hundreds for others. This is a great thing and the Board facilitates it, beyond just the connection between buyers & seller/manf.s.

I thought it might be useful to start a thread on the best way to do such small production runs. That might be helpful to the people who are making these items, and, less directly, help potetntial buyers since it might stimulate more items available and lower prices for the ones that are. Then, if something pans out, Pelcian might want to pick it up (e.g. the big motor badges that are made like the smaller ones -- 3.2 and 3.6).


BTW, I know nothing about this -- as a scientist, I only do things once -- after that no one is interested any more (actually only a dozen people are interested the first time! - but that's another story). I just thought I'd start this to see if it bore any fruitful exchanges between ya'll.

[A Quiet Boom]

I think a thread is a great idea. I made aluminum inserts for thread repair for two board members as well as myself using a manual mill and manual lathe. I could have farmed out the work for a CNC lathe and made hundreds but I didn't find enough interest. The biggest problem with small runs is obviously tooling up. For example at my plant we have a four color Flexo press that I could use to make decals but by the time I bought dies, print plates, print rolls, etc. I'd have well over a grand in tooling. Print plates for example are typically about $100 ea. and even a single color job would require at least a die and sheeter at a cost of about $400. Then there's purchasing roll stock and ink. Even though the time on all these machines is free it's the tooling that kills me. Right now I'm working with RoninLB to make billet watershields but so far I haven't mastered the mill enough to cut the curved portions, on a CNC this would be a piece of cake. Same goes for the motor badges you mentioned, easy on a CNC once programmed but very time consuming on a manual mill. North Coast Cab and I are going to make our own strut braces but that's simple welding and gathering the correct parts. With the time involved (guessing) and a small profit tacked on we wouldn't be to far from the cost of a mass produced item.

I'm attaching some pictures of a hub/brake adapter I made for my brother inlaws TVR 2500M to use Mustang SVT Cobra rear brakes and spindles. It took 20 hours machine time to make this piece (NCC saw the welded prototype when he was down) from a 24lb block of billet. 20 hours for one adapter would be hugely expensive but again cheap if done by computer. I'll just post a few pics and post the rest in OT.

[A Quiet Boom]

I don't want to hijack this thread with the pics of the machinework so please post comments about it here:

Custom adapter for TVR 2500M

Thanks

[jim72911t]

Interesting topic, and I thought I'd chime in since I was indirectly referenced in the original post. (I'm Jim, the Grille Badge Guy )

Without getting into too many boring details, the badges started as a direct response to a thread by Thrown Hammer, who asked if anyone has or would make some 3.0/3.2/3.6 badges for the big bore guys. Since no one, for various reasons, seemed to be able to commit to the project, I thought I'd take on the challenge. As a manufacturing manager in a small, low production, flexible CNC shop, it would be relatively easy for me to come up with a suitable design and secure the machine time to make the badges. I had made numerous pieces for my 911 in the past, so the idea wasn't new.

The plan, at first, was to go in on a weekend and knock out a few dozen badges for the Pelicans. Wasn't so easy, I'm afraid. First, the design work took the better part of two weeks of evenings to come up with a satisfactory font that looked close enough to the P-car script. Then there were issues with manufacturing (prototyping a piece takes a bit of time, $$, and effort, even for something as simple as a grille badge) mounting, finish, packaging, shipping, etc. The final hurdle came when the total number of orders exceeded what I could reasonably do in a weekend, making it necessary to put in a proper purchase order with my boss, which brought along the added financial pressure of having to pay for the badges if people backed out. In essence, what I was doing was developing a product for market.

It was a lot of work, to say the least. To complicate matters, my home computer died the week I was to begin shipping, leaving me wtih a few angry Pelicans wanting to know where their badges were.

In summary, I think that the main reason more of this doesn't happen is becase of the time, money, and effort that goes into small production runs, even for simple parts. See, making the parts is easy. It's all the work that happens before and after that takes the most time. Example: If you came into my shop and asked me to quote a price for a simple, $3 "widget," the total price for one unit would be pretty darn close to the total price for a hundred units. Or, put another way, it's gonna cost $XXX just for me to turn the machine on.

I've since taken it to the next level and have started a small side business selling badges and pedal sets (available through our host), with more on the way. I still work full time at my "real" job, but making a few Porsche parts on the side allows me to pay for (and write off) a few parts here and there for my '72T project.

Hope this helps, and sorry for making you sit through Manufacturing 101

Jim Hollinger
Owner,
JMHP

[A Quiet Boom]

Jim,

I know exactly what you mean. My younger brother is actually considering a CNC/manual mill for his home shop. He's an engineer and know CNC programming very well. His plan is to make parts he needs and then do piece work for small machine shops that don't have CNC machines. With the coating machine and printing press at my shop we have to charge up front for tooling and like you say 1 part costs about the same as hundreds, even thousands in some cases. That's the problem I had with the aluminum case savers, I'd have to buy several hundred to get them at a reasonable price and then hope I sell them all.

[Randy Webb]

Thanks, Jim. That's useful insight. Any ideas you have for reducing the , money, and effort that goes into small production runs, even for simple parts would be welcome, I'm sure. I like my badge BTW (directly from you , with no black on it like the ones thru Pelican now).

Let me just say that I didn't want to limit this thread to machine work only -- CF layups, injection molding (dist. cap rotors for 12 wires), and other manf. techniques are also possible. I even know a guy who forged his own wind tunnel chamber to measure flight energy in birds....

[cegerer]

While we're talking about CNC's and badges, is it possible to replicate the '3D' lettering of the newer Porsches on a CNC machine? I'd like to replace the "Cayenne S" with "955 S" .....

[jim72911t]

Randy: I'm always trying to figure out ways to reduce the amount of time and effort that goes into manufacturing small production runs; heck, I do it all day long! Unfortunately, I couldn't go into detail here; it's a process of continuous improvement, one that takes years of getting it wrong to learn.
If you have specific questions, email me, and I might be able to help.

BTW, I wasn't trying to be too CNC or machine work specific, just rambling about something I know a little bit about.

cegerer:
I'd be afraid to try that script with a CNC and my current manufacturing techniques. Those badges are most likely cast, and it would be cheaper/easier to use that method. I'm not saying that it couldn't be done on a CNC; in fact, I can think of several ways it can be done, but at a significant cost.

[rcwaldo]

Randy this is a great idea for a thread..

I am in the process of starting a small LLC based company for the composite parts I have been making for the Corvette guys. I also have many 993 projects going at the moment. Love to share details to what goes into making composite parts, tooling, drill and trim fixtures etc. Basically the whole process. Never sure if anyone was really interested what happens with this end of the automotive aftermarket?

With a manufacturing background of low production expensive race cars, we have been forced to work smart in order to have any chance at being profitable..

This is the golden question, at what quantity do I tool up for higher production runs??

Chris

[Randy Webb]

"I'm always trying to figure out ways to reduce the amount of time and effort that goes into manufacturing small production runs..."

I meant more in the context of Pelican - e.g. people pledging to buy x so you could show that to your boss, etc.

[Zeke]

I had two prototypes made of items for the 914 when I had mine. Short runs of manual or CNC machining aluminum billet parts came back to me at prices three times higher than the market would bear. I sent the engine item to Jake Raby as a gift. That was a $1000 gift.

But wait, that's nothing.

In 1990, I designed and had a prototype made and then a production die made for a plastic extrusion for a house window product. That one cost me $40,000. The product broke even for a while, but I never saw my investment returned. Eventually, the product was taken off the market due to lack of sales during the recession.

Ask me about short runs and I'll tell you don't quit your day job and plan on having fun, because profits are hard to come by. It might be cheaper to join a country club and play golf. I have learned my lessons.

Then there's people like Chuck Moreland who make out. But, for every CM. there must be 100 Milts.

[Early_S_Man]

Many of these same concerns relate to an electronic project, though the production/assembly technology is different!

The prototype or proof-of-concept version is usually hand-wired, and somewhat of a 'kludge' from an esthetic standpoint.

If a small run of 2-20 is anticipated, a printed circuit board can be designed and fabricated by optical/photographic methods ... iron-on film, UV-sensitive resist, etc. You can't really afford silk-screened PC boards unless you anticipate making 100 or more, as the setup costs will kill you ...

The biggest potential hassle with a large run of silk-screened PC boards is if you run into a supply problem getting components somewhere down the road ... and the change in suppliers means a new component like a relay obsoletes the board design because the pin layouts are different. A minor 'corrections' or engineering revisions can be handled by a few 'patches' with hand-run wires ... but, physical constraints sometimes mean scrapping the original board design. Even Bosch had seven or eight different board layouts in the 1969 - 1977 run of the 3-pin CDI circuit board ... and it is not evident why so many different layouts were used! At least the magnesium casting for the case wasn't changed ... not even when the change to the 6-pin units was made!

[jim72911t]

Randy,
Point taken, and I apologize for the tounge in cheek reponse. When I first started making pedals (my first product, actually), I did sit down with my boss and discuss potential sales numbers, and he acted as sort of a partner for a year or so. Since then, I've taken the business out on my own, so I have to analyze potential return on investment and decide if a product is worth tooling up for.

As far as people pledging to buy a proposed product, whether it be on this board or elsewhere, it has been my experience that many people will express interest, but quite a few less will actually follow through and purchase said product. Not complaining, just stating my experience. With the badges, I'm still sitting on a few custom badges that I've made (and paid for), where the interested buyer has never come through with payment. Cost of doing business, AFAIK, and the badges will eventually sell, albeit at discount prices.

I think the safest way to keep from manufacturing unsellable, one off or low production parts is to at least require a deposit before the parts are made. That way, you might at least keep from losing money on the deal.

I haven't resorted to that, because, for the most part, the people I deal with in the Porsche community have come through, and appreciate the product.

Jim

[kstar]

Here's an interesting story germane to this thread:

http://story.news.yahoo.com/news?tmpl=story&cid=528&e=2&u=/ap/20040818/ap_on_hi_te/online_manufacturing

Excerpt:

It's the Internet Revolution meets the Industrial Revolution: a new program that lets people design 3-D objects like car parts and door knobs in metal or plastic then order them online.

Programs for computer-aided design, or CAD, have been around for decades, but eMachineShop.com appears to be the first service that checks whether a design can be made, tells the customer how much it will cost. If the customer wants the item the design goes to a "real world" machine shop for manufacturing.

FYI.

Best,

Kurt

[RickM]

I have some of experience in small lot runs of machined or stamped parts/products. Over the years I've found a few businesses that are willing to, or specialize in, this mode of production. Additionally, I am usually successfull in sourcing products or work overseas. The most familiar may be the sourcing of the Phillips light bulb bases for the 911 LED guage project....sadly defunct now.

I can also cast a variety of parts in Antimony or Pewter (Aluminum is next). The scripts or Steering wheel badges would be relatively easy and inexpensive. However, because of copyright laws one would never be able to profit enough from the effort.

If anyone needs help with any of the above I'd be happy to help. Great idea for this thread.

[mskala]

Cool thread. I can agree mostly with Warren about the cost of
short-run electronic projects. Although the improved outsource
manufacturing processes and competition via the internet has
made it very easy to build, and cost-out the items beforehand; lots of PCB builders and component suppliers.

But it would be completely impossible to compete with large
companies' large runs. You need a product nobody else wants
to make. I look at stuff that sells for <$20 and think somebody
in China had to produce a million of these! And fast before the
next fad hits.

I could give anybody leads or help in making that kind of stuff.
Right now I am nearly done with the procurement of short run
of parts for a track-car picture-in-picture processor I hope to at
least sell to break even, starting with the various BBS's.
(www.rcd-video.com)

Mark S.
'70 914-6

[project935]

Great thread.

I've had some ideas of replicating 935-style suspension parts and selling for less than what is out on the market today. I haven't sat down and done all the number crunching yet, but something tells me what is currently on the market is greatly over priced.

If anyone needs CAD/CAE assistance with any of their ideas, drop me a line ... I'd be happy to help. Same goes for reverse engineering anything - I do it all the time. Eventually I hope to have my entire Porsche project in CAD.

[midlife]

Well I just joined the ranks of the low volume Pcar parts manufacturers. See my post:Check out these knobs! (dash knobs)

As a hobby, or to help justify making parts for yourself it is a great idea. As a business it is a tough to make money in low volumes. Without a lot of tooling, piece cost is high, so the market must willing to accept the price. Today's advancements in Rapid Prototyping, and Rapid Tooling, have made it much faster to get low volume parts, but not cheaper.

At my day job, we design and manufacture automotive components. I try and avoid the low volume parts. With all the quality requirements of the industry even the simplest of parts requires a huge investment. There are DFMEA's, PFMEA's, Control plans, DVP&R's, PPAP's, capability studies, and on and on.
That is the reason I enjoy working in my own shop. It is satisfying to just design and build something without a bunch of meetings, and a mountain of paperwork.

[RickM]

Here'a perfect candidate for this thread: Oil Sender Custom Tool

[BURN-BROS]

Good input!
Knowing your target demographic is very important. Zeke found out that the 914 crowd generally does not spend the money so there are only few vendors and products out there for them.

Try to do as much as you can inhouse. It is truly the only way to make short runs happen.

The company that I work for is getting a 3-d printer tomorrow and a 12' x 12' x 6' 5-axis router next week. We have moved into prototyping and making a customer's idea a reality.

Aaron