Tuesday, February 27, 2007

Chapter 11. Make good mold choices





Skip this chapter at your peril.

Design refinements

A

s I mentioned earlier, a business-savvy technical manager called Brad Raker working for Cascade Plastics scrubbed my designs for ideal material use and structural fit. I didn’t know at the time that I needed that help, but it would have been a disaster had he not stepped in to bring my designs to a workable condition.

  • Unless you are an expert in the field, a mold-making expert will need to refine your designs before you try to make molds.

You might not know this, but besides the cost of making the thousands of pieces later on, you the designer have to pay for the necessary molds that make those pieces. This is an up-front, sunk cost. It is an unavoidable part of the investment you will make, long before you secure a customer, to get your product to market.

You might think, “With my wonderful product idea, my manufacturer will be happy to cover the cost of mold creation knowing that they’ll make lots of money downstream” or “My manufacturer will amortize the cost of the molds across many iterations of production”.

Not true on either count.

You must budget for the cost of mold creation. Most mold creators that I have spoken with ask for a 50% down payment before they touch the job and the other 50% on the day the molds are completed. So, let us say your molds cost $120k; you will have to write a check for $60,000 to get them to start the job. Then, when you and the mold builder are both satisfied the sample injection parts they create with the new molds are adequate, you both sign off and you write another check for the remaining $60,000. You still have no salable product, of course.

  • You must pay up-front for the molds.

Three main aspects of mold creation have an enormous affect on the complexity and cost of the mold:

· Number of undercuts

· Number of cavities

· Greatest thickness of part

There are other factors, such as material, color and additives, but those three are the most significant in almost every case.

Number of undercuts

  • Undercuts significantly increase the cost of the mold.

An undercut in a plastic part results in a protrusion inside the mold that sticks into the plastic piece when it is in the mold, and would usually make it difficult or impossible to extract the cooled, solid piece of plastic from that mold. Again, think of a piece of solidified jelly which would not come loose from the mold if there were parts of the mold jutting into the side of the jelly to stop it from moving out of the mold. Jelly can squeeze and stretch to allow you to unhook it from within the jelly mold if that were necessary, but plastic is not so forgiving. Most plastic pieces are a lot stiffer than jelly, and will resist efforts to extract them from the metal mold if even a small protrusion in the mold gets in their way.

Take the time to go look at a jelly mold. Look directly into the mold and you will see every centimeter of the inside of the mold from your “bird’s-eye view”. There are no areas inside the mold that are “hidden” from sight. Thus, there are no undercuts in the jelly mold. The best molds, for plastic injection, are those that have no undercuts, and thus, do not need parts that move out of the way for you to extract the plastic from the mold. Molds with no undercuts are less costly to buy, less complicated to operate, last longer, and do not break as often.

To make a mold with no undercuts, you often need to do some clever design work. Experienced CAD designers look for ways to lessen or eliminate the need for undercuts from the beginning of the design process. Plastic products are everywhere, and once you understand the nature of the undercut, you will notice many pieces of plastic that were carefully designed to avoid supporting undercuts in the mold.

Next time you get a chance, take a look at the common construction hat, pictured in Figure 9. The mold for this construction hat was made without the use of moving parts because there are no undercuts designed into the hat itself. Take the straps and other attachments off a standard construction hat, turn it upside-down, and you have an excellent jelly mold.

There are other ways of making plastic products, like blow molding, which is used for products like soda pop bottles, but the undercut issue remains a challenge for those of us who wish to make injection molded parts.

Figure 9 – the common construction hat (without attachments)

The design represented in Figure 5 on page 61 is a mold creation nightmare. It is theoretically possible to make the mold for it, but it would probably be well over $100,000 even for a single-cavity mold to make that one piece.

Dave Mesaros (Cascade Plastics) gave me many tips on how to design without undercuts, so I went away and redesigned that piece without them. It was a challenge, but I had come a long way in CAD proficiency so I felt I was up to the job.

I redesigned the complete half-hexagon from scratch, this time, without any undercuts, represented by the illustration in Figure 10, which also includes the “sculpting out” of unnecessary plastic thickness done by Brad Raker, who also made a bunch of other subtle refinements to the piece. The sculpting, by the way, reduced the material consumption, made the product lighter and quickened the production time, all without compromising the strength or function of the piece.

That sculpting work took Raker a few hours and is another example of why you need expert help at this stage.

Figure 10 – redesigned without undercuts

Critical: Number of cavities

Imagine yourself in the kitchen again. You are using a baking tray containing twelve identical cavities (see Figure 11 on page 76). You heat up the oven, pour your mix into the twelve cavities and bake all twelve muffins at the same time.

Now, imagine how long it would take if your baking tray had only one cavity! It would take twelve times as long to bake and cost twelve times as much to perform that part of the operation. An experienced baker will remind you that setup and cleanup also take time and therefore have a cost.

When you bake twelve muffins at a go, I estimate the cost of the ingredients is about 50% of the total cost of the muffins. Let us guess at a price: Baking twelve muffins in a twelve-cavity baking tray costs $2.40, so each muffin costs 20 cents: 10 cents for the ingredients and 10 cents for its share of the oven time (1/12 of the $1.20 it costs to run your oven for an hour).

With two adults and four kids in the house, that might equate to two muffins each. In my house, muffins disappear so fast, they don’t have time to cool fully.

Imagine once again that your muffin tray had only a single cavity. Your family members would still want two muffins each, but now, you must bake one muffin at a time. The baking part of the operation takes a whopping twelve hours instead of one hour, and the overall cost is now $15.60 ($14.40 for the electricity and $1.20 for the ingredients) for the same twelve muffins. This equates to $1.30 a muffin, or about thirteen times the cost of the ingredients. That’s just for the electricity and materials and doesn’t take your time into account.

Thus, a single cavity baking tray means each muffin costs $1.20, and a twelve-cavity baking tray brings the cost down to 20 cents a muffin, or one sixth of the cost of when you bake only one at a time.

You saved some money by buying a single-cavity baking tray, perhaps paying $5 instead of $20, but you must pay a lot more per muffin come baking time.

Figure 11 – a baking tray for 12 muffins

Figure 12 – a baking tray for 1 muffin

With respect to cost, the mathematics of mold creation and plastic injection are very similar to that of baking trays and muffin baking.

When you request a quote for mold making and the corresponding plastic injected pieces, ask for two sets of quotes: one set for a low-cavity count, and the other set for the highest cavity count they can make in their shop. Do not mix them up. The first set of quotes, which are for the low cavity count, will have a low price for the molds and a high price for the plastic injected pieces. The second set of quotes, for the high cavity count, will have a higher price for the molds and a (probably) significantly lower price for the plastic injected pieces.

Remember too, that you may have a product consisting of large quantities of one part and small quantities of another. If you were making plastic knives, forks and spoons, you might plan to make two forks for every knife and ten forks for every spoon. So, each mold you get made might not have the same optimal cavity count.

  • Low cavity count means cheaper molds but more expensive parts.
  • High cavity count means more expensive molds but cheaper parts.

When I was ordering my own molds, I had misunderstood the significance of cavity count. Or rather, I did not pay enough attention to the downstream production costs when I was looking at the cost of the initial molds. Perhaps it was because I was a novice, but I was not looking for a huge difference and I did not focus at all on the cavity count. Thus, I ordered molds with low cavity counts without understanding the entire significance. As it turned out, I got lucky because three of the six molds were to need improvements anyway, and they alone accounted for 90% of the production cost of the entire product. So, in the end it did not make a difference, but had it been the cheaper three molds that needed the improvements, the mistake would have cost me about $30k.

  • A good rule-of-thumb for minimum plastic piece price is about 150% of the material cost.

Greatest thickness of the part

Plastic parts made in plastic injection molds are made by injecting hot resin under pressure into the molds, and then cooled while still in the mold. After cooling, the mold is opened, and “pins” force the piece of plastic to pop out of the mold. The mold is closed again and the process is repeated. Although technology has made it possible to make the process more reliable and efficient over the years, plastic injection molding technology has not changed in decades.

Some parts take a lot longer to cool than others. The thickest part of the plastic piece will decide the minimum amount of time the pieces need to remain in the mold for cooling. How do you measure “maximum thickness”?

In Ireland, where I grew up, nowhere on the entire island is more than 50 miles from the coast, despite the country covering about 30,000 square miles.

Imagine for a moment a piece of plastic where the furthest molecule of plastic might be from the surface of that piece of plastic. For example, consider a plastic gardening trowel. Its solid plastic handle contains its thickest section at 1.5” wide, so the furthest molecule of plastic might be about .75” from the surface. During manufacturing of that part, it might take an entire minute for that piece to cool enough for the mold to be opened. That will double the machine time element of your costs.

  • Cooling time in the mold is expensive

An irrelevant subtlety to the untrained eye, perhaps, but a simple adjustment in the handle to make it hollow or of a different shape will shorten the cooling time during manufacturing and cause a significant decrease in manufacturing cost. A solid plastic handle might cost several times more to manufacture than a hollow handle, not just because it has more plastic in it, but because it ties up the mold for a lot longer while it is cooling.

  • Thicker pieces of plastic parts significantly increase the cost of manufacture because they take longer to cool.

Summary – the simpler the better

I cannot stress enough how beneficial it is to have parts with no undercuts. Take the time to design your parts so they do not have any. Many products on the market have been designed so individual parts are injected molded separately, perhaps even in the same mold and then snapped or glued together by an assembly line worker or end user after the fact.

Some plastic materials are more amenable to being glued than others. Some plastics bond to other materials reluctantly.

Whatever way you design and build your product, always, always be on the lookout for a simpler way to construct it.

  • Every little simplification you make to your product today means less product risk tomorrow.

Obvious as this may sound, it is worth repeating: As design complexity increases arithmetically, the risk of failure increases geometrically.

  • Product risk increases geometrically to the rate as product complexity increases arithmetically.

A low cavity count, undercuts and a thick section of your plastic part each increases part cost dramatically. Here is a rough formula that you can use to see just how much of a difference a slight change in the design can make:

$0.10 + ($100 x M x MC x (1 + (U/5)) * T / CC)

M = Mass (lbs)
MC = Material Cost (per lb)
U = number of undercuts
T = max Thickness
CC = Cavity Count

Let’s look at some hypothetical examples:

You want to manufacture a special plastic handle you have designed for a garden fork. It lessens the risk of RSI[1] and it attaches to any standard garden hand held tool. In the table illustrated in Figure 13 on page 81, each significant variable is changed to see what affect it has on the individual unit price during manufacturing. The variables are changed, and the resulting unit cost is calculated, for each of four different situations, each represented by a single row in the table (Figure 13). Our CAD program told us the spoon we designed should weigh about 0.05 lb.

The results in Figure 13 are what you might expect to pay a manufacturer in the United States (assuming 2006 material prices) for each spoon. I remind you that this formula is rough, reflecting only my own experience, and that you need to look at the specific quotes provided to you by the manufacturer you are working with. This formula provides a general impression of how significant these factors are, not a means to calculate your expected costs for planning purposes.

Do not forget to check your own estimates with people who make molds and injected plastic parts for a living. My formula is just a way of showing how costs can spike because of the tiniest adjustment in your product design.

Mass
(lbs)

Material
$/lb

# undercuts

Cavity count

Max thick

Cost/part

.05

$1.50

0

16

.2

$0.16

.05

$1.50

0

2

.2

$0.85

.05

$1.50

4

2

.2

$1.45

.05

$1.50

4

2

.4

$2.80

Figure 13 – theoretical costs of manufacturing

Clearly, if you plan to make large numbers of your parts, anything you can do to increase cavity count, decrease maximum thickness and eliminate undercuts will lower your costs significantly. In the theoretical example above, a piece made with four undercuts, two cavities and only 2x the thickness can cost sixteen times as much.

However, if you plan to make small numbers of parts over the lifetime of the mold, the same rules do not apply. Because the cost of a mold is independent of how many parts you make with it, if few injected parts are made with it, the per-piece cost is greater. When millions of parts are made with the same mold, the cost of the mold to make them is shared over so many pieces the mold price per injected part may get close to nothing. For example, if a single mold cost $25,000 to make, and it was used it to make five million parts, each part is assigned one half of one cent to cover its share of the cost of building the mold. In fact, when you know you are going to make millions of units with your mold, it is wise to pay to get it done right. Paying an extra $5,000 on top of, say $25,000, to get that extra expert care to refine it will be worth it. It might add a 10th of a cent to the cost of each part, but might make that part significantly superior in quality.

Figure 14 – expensive night-vision binoculars

On the other hand, consider this low-quantity production example: You paid $25,000 to build a mold to make the casing for night-vision binoculars, pictured in Figure 14, of which you expect to sell 2,000 units. Each casing would take $12.50 as its share of the mold cost. The entire production run could probably be completed in one day, even with a single cavity mold. If you were to make a mold with four cavities, the unit manufacturing cost would increase when you add the mold cost share to each unit. This is because the mold will not make many units over which to spread the increased mold cost.

  • Small productions numbers suggest lower cavity counts. Large production numbers suggest higher cavity counts.

That may sound obvious, but novices get that wrong all the time. The point here is to get expert help. With all the stupid questions I asked and “free advice” I clearly needed, I’m sure I frustrated several of the manufacturers I was considering. One manufacturer declined to give me a quote in the end. He probably saw a future of having to handhold me through the process while I learned everything at his expense and thought the better of it.

Dave Mesaros and Brad Raker in Cascade Plastics were interested in my product and clearly had the experience of dealing with other entrepreneurs. I felt like they were partners in trying to make own my business successful, so I signed them up to do the design scrubs, molds and final production.

Cost of this stage: $0. Costs so far: $30,800

End of chapter exercise

  • Go into your kitchen and find a jelly mold. Take a hard look at it. Notice how it is made such that solidified jelly will slide out of it easily. If you have time, go through the effort of making some jelly. When you remove it from the mold, take note of how easily that happens.



[1] RSI: Repetitive Strain Injury – physical injury that occurs by doing the same physical activity over and over.

No comments: