PDA

View Full Version : How much would this cost?



SRT Mike
04-26-2007, 11:20 PM
I sort-of asked this before.

But..

Imagine a motorcycle tail light. It has a black ABS plastic back plate, and a clear acrylic front plate. Sizes are roughly the size of your hand (in general). Some of them have fancy fresnel lens inserts - assume those are not necessary.

What sort of cost would be involved in taking an OEM factory tail light and reproducing it? The idea is that the OEM light uses bulbs and the aftermarket part will be LED, with a custom circuit board installed. The OEM light would be slightly changed from OEM (i.e. maybe moving the mounting bosses to facilitate mounting of a circuit board).

What sort of cost/volume is involved here?

I know the OEM's use things like carbide square rods precision ground and bolted together to create the textured lens surfaces with great precision. But consider much of the Chinese import stuff - they are not nearly so precise. Let's say the mold doesn't need to be able to run more than a few thousand parts (maybe 6061 or 7075 AL?). All the light lenses are straight pull molds.

What would the price range be to get such a mold made, and what volume an cost-per-part would we be looking at?

I have heard mold prices from $2k to $40k, per-part costs from $0.30 to $5.00 and volume req's from 4k to 100k units. As a datapoint, there are other companies selling reproduction OEM lenses for $35, with a circuit board installed. This is the retail price, meaning dealer price must be more like $20, and the OEM is making money at $20 (after shipping), so I figure they can't be into the lens for more than a few bucks.

I'm looking to get into a price point of hopefully $5 for the front and back of the lens. And I am looking not to front many thousands up front. And I'm hoping to buy maybe 500-2000 parts a year over a product life of 4-5 years. I know that may sound 'off' to some, but, 2-3 other companies are already doing this - so if it's not possible, how are they doing it?

I have no molding experience, but we have plenty of machining experience. What would the feasibility be of buying a molding machine and learning to do this ourselves? I know the principles of mold design (drafts, runners, sprues, part thickness, bosses, gussets, etc) but I've never machined a mold or run a molding machine. Would it take me years to get up to speed with a mold, or could I possibly machine my own mold out of AL and be molding parts in a matter of months?

Thanks - I apologize for being a total n00bie when it comes to this subject.

turmite
04-27-2007, 08:57 AM
Hi Mike,
My first suggestion would be to proof the part, then have a 3d model made or make it yourself if you have 3d cad experience. I understand your desire to get a target price before doing the needed work as I have been in the same boat many times, and is the reason I feel qualified to give you some direction.

There are small mold companies out there that will work with you but not in hundreds of parts at a time. They can run several hundred parts in a few hours and it is simply not worth their time for the set up. I hate that phrase, but after having experience with it myself, I do understand it better.(nuts)

Have you proofed the idea?
Can you provide the 3d model?

Answer those two questions before you go on and it will make the process a little easier.

Mike

mc_n_g
04-27-2007, 06:06 PM
If you have something already I would try one of the rapid prototype places first to see what type of cost there is and feasability.
Right now you have too many variables. Too many possible changes which cost money.
mc_n_g

InspirationTool
04-27-2007, 06:50 PM
Can you machine them instead, and then vibratory finish them to polish them?

-Jeff

Kipper
04-28-2007, 04:51 AM
I've been thinking of replacing the entire rear light on my motorcycle for some time and think it will cost me roughly £10 to do it. Fortunately I have materials left over from many years ago when I made lights for some 1/2 scale cars I made for my kids. I must see if I can find some pictures :beer: My method was to use "vinamold" and UV stabilised resin and hand make the mould....Oh how times have changed...I can CNC the mould's now :yay:

SRT Mike
04-29-2007, 10:56 PM
Here are some pics of a particular motorcycle tail light I want to replicate.

The back

http://www.hightechspeed.com/images/web/1.jpg



The front

http://www.hightechspeed.com/images/web/2.jpg


Iso view of front

http://www.hightechspeed.com/images/web/3.jpg



To answer specifics:

Turmite, the thing is - these are already molded parts - so things like draft angles, thicknesses, etc have already been designed in. I have not modeled the part, because it's a relatively complex 3D shape, and it would take almost no time to laser scan it, whereas if I put it on the machine and uses the probe head, it would take hours if not days and whomever was gonna mold it would be re-doing it anyway :)


mc_n_g - I thought the variables were pretty constrained? I dunno, maybe I am wrong. In the case of the pics I posted, I want exactly what is shown. There are a few cases where instead of a mounting boss in one place, I want it in another place, or instead of a tab that slides through a board, I want a mounting boss. But nothing that would complicate the mold design (actually everything I'd change would make it easier).


Inspiration - the problem with that is firstly that vibratory finishing them would round all the edges, which will make the lens look like crap (i.e. won't do any diffusion really). Secondly, the light is maybe 5" wide, 4" high, and about 5" deep. So we'd be talking a BIG chunk of acrylic to start with, and many hours of machining. And I can't even imagine machining a few thousand of 'em :)




I was wondering if there was some acrylic casting process, although I don't think there is. I know you can cast resin, but not with the clarity I need. I also had considered thermoforming. Maybe taking a flat sheet of 1/8" acrylic, and heating it up until it's pliable, then putting it on a positive image of the inside of that tail light, letting it "droop" over it, then clamping on a top section such that the cavity between is exactly what I want my finishe part to be. But I figure if it was as easy as that, then such parts as above would not be injection molded.

Thing is - the fact remains that various manufacturers make these. I used to sell some and I know the market is a thousand a year MAYBE on a hot seller, and the products run from 2 to 4 years in production. Other companies sell these parts for a price that lets me know there is only a few bucks in the plastic parts at most. So I'm wondering if I can approach a price of perhaps $10/part (part meaning clear front and plastic back) by having someone do the molding for me? Or if I would have to do the molding in-house to get that level of price?

Same with machining I guess... we have a lot of AL parts we make. If I had to outsource them, I could never make them at a cost that allows me to sell them on to distributors. But if I make them in-house, we do the fixtures and set-up, spend $3 in raw AL and can cost the part at $5 with machine time. If I was getting them made by someone else, they'd be $15-20/part or more, depending on how many I needed.

So, with the info above, can anyone guesstimate a ballpark cost to do a mold, an a per-part cost (on the acrylic lens + ABS back plate)??? Or is this something I'm gonna have to figure out how to do in-house (injection molding)?

Thanks!

mc_n_g
04-30-2007, 07:02 PM
Holy $hit that is a nasty part to redesign from scratch.
The little angles and curved surfaces are not nice at all. To redo without a CAD drawing or a full scan would not be fun. The little scallops on the bottom are not nice either. That is at least a 3 piece mold. It might be doable in 2 but tough parting line with the pattern in the lens.
The amount of polish for the surface for acrylic is nuts with all the little angles. Definite need for chromed surface.
If you are truely serious I would try to get someone to laserscan it to get a point of clouds model to start from.
I would try to cast one from a silicone mould and use pourable clear resins with vacuum to pull out the bubbles. You might be surprised.
The problem is there is no way to mass produce the mold. You have to keep on remaking the silicone to generate a few molds.
You are going to have to make a lot to recover teh cost of the mold if you go the machining way. The price of plastic is not the problem. It is the cost of the design, mold and polishing (possible chrome cost too). With acrylic the surface finish will drive your clarity. Small swirls will form from non-smooth surfaces cause ripples in the intial plastic skin.

mc_n_g

turmite
05-02-2007, 12:33 AM
To answer specifics:

Turmite, the thing is - these are already molded parts - so things like draft angles, thicknesses, etc have already been designed in. I have not modeled the part, because it's a relatively complex 3D shape, and it would take almost no time to laser scan it, whereas if I put it on the machine and uses the probe head, it would take hours if not days and whomever was gonna mold it would be re-doing it anyway :) Mike if you do a search in the RFQ/Employment forum here on the zone there is a company that posted late last year/eearly this year that was a scanning company. To scan and provide a stl was not all that bad for something much larger than your part. Problem is you part is actually two so....... Anyway if you can't find them let me know and I will try to dig up the email, assuming I still have it.


mc_n_g - I thought the variables were pretty constrained? I dunno, maybe I am wrong. In the case of the pics I posted, I want exactly what is shown. There are a few cases where instead of a mounting boss in one place, I want it in another place, or instead of a tab that slides through a board, I want a mounting boss. But nothing that would complicate the mold design (actually everything I'd change would make it easier).


Inspiration - the problem with that is firstly that vibratory finishing them would round all the edges, which will make the lens look like crap (i.e. won't do any diffusion really). Secondly, the light is maybe 5" wide, 4" high, and about 5" deep. So we'd be talking a BIG chunk of acrylic to start with, and many hours of machining. And I can't even imagine machining a few thousand of 'em :)




I was wondering if there was some acrylic casting process, although I don't think there is. I know you can cast resin, but not with the clarity I need. I also had considered thermoforming. Maybe taking a flat sheet of 1/8" acrylic, and heating it up until it's pliable, then putting it on a positive image of the inside of that tail light, letting it "droop" over it, then clamping on a top section such that the cavity between is exactly what I want my finishe part to be. But I figure if it was as easy as that, then such parts as above would not be injection molded.

Thing is - the fact remains that various manufacturers make these. I used to sell some and I know the market is a thousand a year MAYBE on a hot seller, and the products run from 2 to 4 years in production. Other companies sell these parts for a price that lets me know there is only a few bucks in the plastic parts at most. So I'm wondering if I can approach a price of perhaps $10/part (part meaning clear front and plastic back) by having someone do the molding for me? Or if I would have to do the molding in-house to get that level of price?

Same with machining I guess... we have a lot of AL parts we make. If I had to outsource them, I could never make them at a cost that allows me to sell them on to distributors. But if I make them in-house, we do the fixtures and set-up, spend $3 in raw AL and can cost the part at $5 with machine time. If I was getting them made by someone else, they'd be $15-20/part or more, depending on how many I needed.

So, with the info above, can anyone guesstimate a ballpark cost to do a mold, an a per-part cost (on the acrylic lens + ABS back plate)??? Or is this something I'm gonna have to figure out how to do in-house (injection molding)? If you have in-house injection molding this is going to be your least expensive option, aside from the mold cost of course.

Thanks!

Mike

ltmquik
05-04-2007, 04:29 PM
Your best bet for the quantity, 1000 pcs per year and a six (6) year life, would be an aluminum injection mold, two (2) cavity (1+1 family) mold with a runner shutoff (for the two different materials).

By my quick guess you would be looking at around $9,500.00 US in mold costs. The parts cost would be around $10.00 for the two parts.

If you want I can have my plastic injection molding company (Plastic Model Engineering, Inc.) provide you with a formal quote. I will need more detailed information and maybe a sample part.

Thank you.

Kipper
05-06-2007, 07:56 PM
Here are some pics of a particular motorcycle tail light I want to replicate.

The back

http://www.hightechspeed.com/images/web/1.jpg



The front

http://www.hightechspeed.com/images/web/2.jpg


Iso view of front

http://www.hightechspeed.com/images/web/3.jpg



To answer specifics:

Turmite, the thing is - these are already molded parts - so things like draft angles, thicknesses, etc have already been designed in. I have not modeled the part, because it's a relatively complex 3D shape, and it would take almost no time to laser scan it, whereas if I put it on the machine and uses the probe head, it would take hours if not days and whomever was gonna mold it would be re-doing it anyway :)


mc_n_g - I thought the variables were pretty constrained? I dunno, maybe I am wrong. In the case of the pics I posted, I want exactly what is shown. There are a few cases where instead of a mounting boss in one place, I want it in another place, or instead of a tab that slides through a board, I want a mounting boss. But nothing that would complicate the mold design (actually everything I'd change would make it easier).


Inspiration - the problem with that is firstly that vibratory finishing them would round all the edges, which will make the lens look like crap (i.e. won't do any diffusion really). Secondly, the light is maybe 5" wide, 4" high, and about 5" deep. So we'd be talking a BIG chunk of acrylic to start with, and many hours of machining. And I can't even imagine machining a few thousand of 'em :)




I was wondering if there was some acrylic casting process, although I don't think there is. I know you can cast resin, but not with the clarity I need. I also had considered thermoforming. Maybe taking a flat sheet of 1/8" acrylic, and heating it up until it's pliable, then putting it on a positive image of the inside of that tail light, letting it "droop" over it, then clamping on a top section such that the cavity between is exactly what I want my finishe part to be. But I figure if it was as easy as that, then such parts as above would not be injection molded.

Thing is - the fact remains that various manufacturers make these. I used to sell some and I know the market is a thousand a year MAYBE on a hot seller, and the products run from 2 to 4 years in production. Other companies sell these parts for a price that lets me know there is only a few bucks in the plastic parts at most. So I'm wondering if I can approach a price of perhaps $10/part (part meaning clear front and plastic back) by having someone do the molding for me? Or if I would have to do the molding in-house to get that level of price?

Same with machining I guess... we have a lot of AL parts we make. If I had to outsource them, I could never make them at a cost that allows me to sell them on to distributors. But if I make them in-house, we do the fixtures and set-up, spend $3 in raw AL and can cost the part at $5 with machine time. If I was getting them made by someone else, they'd be $15-20/part or more, depending on how many I needed.

So, with the info above, can anyone guesstimate a ballpark cost to do a mold, an a per-part cost (on the acrylic lens + ABS back plate)??? Or is this something I'm gonna have to figure out how to do in-house (injection molding)?

Thanks! You "know" that? This is my son at age 5...He cant fit into this any more as being 6'4" it's a trifle hard :) but the lights are still crystal clear as per the manufacturers statement "Crystal clear UV resistant resin will not cloud or crack" 1928 SSK "toy" car sold for a tidy sum 5 years ago. I may add that it was all handmade from steel and wood and travelled 20 miles before recharge.

dansutula
05-06-2007, 10:25 PM
This is right up their ally! They can do this cheaper and faster than anyone in the industry. They have made at least a 1/2 dozen molds/parts for me, and I've never been disappointed and have often been amazed by their process. You will however, need a CAD model to upload to their automated quote process. You may want to find someone with a Nextengine type scanner as a start to building your CAD model, or a good reverse-engineer. 1,000-10,000 parts should be no problem in an aluminum mold with un-filled resin. The downside with ProtoMold is that the per part cost will never go down that low. Your part might cost $4-$6 or more even in moderate volume.
Good Luck!
Dan

ltmquik
05-07-2007, 11:17 AM
This is right up their ally! They can do this cheaper and faster than anyone in the industry. They have made at least a 1/2 dozen molds/parts for me, and I've never been disappointed and have often been amazed by their process. You will however, need a CAD model to upload to their automated quote process. You may want to find someone with a Nextengine type scanner as a start to building your CAD model, or a good reverse-engineer. 1,000-10,000 parts should be no problem in an aluminum mold with un-filled resin. The downside with ProtoMold is that the per part cost will never go down that low. Your part might cost $4-$6 or more even in moderate volume.
Good Luck!
Dan

We provide the same service as ProtoMold but offer guarantees on our tooling.

MrWild
05-07-2007, 12:42 PM
Where does Patent infringement come into play? The lens shape took development and the design may be patented. Why are you trying to replicate it anyway? The assembly looks thin so is it already an LED item? If it is powered by a filiment bulb, the lens design is made for a single point light source. Going to a multiple LED light source makes the lens design a bit worthless for your application. It'd be better to make lenses over each LED diffusing the LED(s) light.

MauricioB
07-31-2007, 10:09 AM
Mike

We offer a free application to calculate the cost of plastic parts. You can fin it at http://www.injecneering.com and it is really absolutely free.

The calculations consider most factors affecting the cost to produce an injection molded part such as cost of virgin material, regrind, colorant, tool cost, machine cost, operator, amortization. The application in called PX1 Cost estimator. I does not give you the cost of the tool or the machine though. You have to enter them, but it will help you analyze different scenarios to cost your product

Hope it helps,

Mauricio Benavides

raynjer
06-23-2008, 02:44 PM
Probably much too late for this project, but laser scanning the part is probably cheapest way to produce the model accurately. Another option is to make an electroformed nickel cast of the existing plastic part which can then be used as an insert in a mold. This does not come cheap, but it gives you a ready to use mold insert that is an exact replica.
I can provide more information if necessary for scanning or electroforming of plastic parts.

plasticpeter
06-29-2008, 04:24 PM
The surface of a lens for automotive lighting is an extremely complex, mathematically-derived shape. They are designed to spread the light exactly where it is needed and away from where it isn't - and this is goverened by some very complex legislation. I also believe that each design needs to be tested and certified before it can be legally fitted to a vehicle.

From experience, I personally doubt that laser scanning will give you enough information to reproduce the functionality of the lens. I could ask my father about this as he was involved in this industry all of his working life. I have seen some of the optical calculations he used to do and it was scary!

plasticpeter
06-29-2008, 04:42 PM
Notwithstanding what I have just said about reverse engineering the product, I agree with what others have said. You may have the machining capability to produce the mould and you may be able to buy-in the processing capability/expertise but you need help with the mould design. It is never straightforward and this is no DIY job.

At the end of the day, use a sub-contractor to design and make your mould. They will also be able to help you with the reverse engineering bit and all of your tooling AND part quoting requirements. And yes, 6000 or 7000 series aluminium alloy is 100% feasible for this project.

I also seriously doubt that you will ever be able to mould it cheaper than a sub-contractor at the kind of volumes you are talking about. ProtoMould will do the whole job for you but may not be as cheap as the guy offering you his services below! That's a good price if he can guarantee end quality (but I don't think he includes the cost of the reverse engineering process).

Lastly, I certainly think there are legal questions arising from this. However you look at it, the design of that part is somebody else's intellectual property and it surely will be protected in some way!!

Mike Stevenson
06-29-2008, 06:54 PM
That is a seriously complicated and expensive mold to make in my opinion.

PEU
06-29-2008, 09:05 PM
Let me focus on the electronics of your project, what kind of LED are you planning to use? 5mm common LEDs or the new breed of high power LEDs such as Cree, Luxeon, Seoul Semi and others?

If you plan to go with the high power ones I would redesign the back of the fixture to be of aluminium and work as a heatsink since these LEDs need a proper heat path.

Also LEDs are different light sources that incandescent, so the reflector you are trying to copy will not work the same way.

Maybe your competitors are able to move low quantities/year by making molds for multiple kind of lens, the mold is bigger and more expensive, but its less than the sum of individual molds.

If you want to design lenses, google: tracepro

good luck


Pablo

Mike Stevenson
06-29-2008, 09:12 PM
Pablo,

I like ribs too. :D

PEU
06-30-2008, 09:07 AM
Pablo,

I like ribs too. :D

Then you should come to Argentina, its paradise for grilled meat :)

SRT Mike
06-30-2008, 09:47 AM
This is sort of an old thread, but I got a lot of great info and I wanted to address a few of the questions raised, especially in the last couple of weeks :)


Where does Patent infringement come into play? The lens shape took development and the design may be patented. Why are you trying to replicate it anyway? The assembly looks thin so is it already an LED item? If it is powered by a filiment bulb, the lens design is made for a single point light source. Going to a multiple LED light source makes the lens design a bit worthless for your application. It'd be better to make lenses over each LED diffusing the LED(s) light.

No issues with patent infringement. I am not aware of any light design that is patented, neither utility nor design patent. And, when you think about it... if there were IP issues, then how could an aftermarket company make non-OEM parts like reproduction wheels, windshields, bumpers, lights, etc.

As for why trying to replicate, because we want to offer additional features beyond the OEM part. The lens design is applicable because the light I posted is LED so the lens design would carry over. In cases where you go from incandescent to LED it's generally preferred to change the lens design but that's not a huge big deal - the main consideration is the shape/contours of the actual lens. Most OEM LED lights have 2 lenses, a fresnel lens over the actual LED and then a diffusion lens on the rear surface. You can take the shape of the light surface and just modify the lens(es) from prismatic (good for incandescent bulbs) to a big fresnel lens or discrete fresnel lenses.

plasticpeter
06-30-2008, 01:40 PM
I don't understand why you now say:

"As for why trying to replicate, because we want to offer additional features beyond the OEM part. The lens design is applicable because the light I posted is LED so the lens design would carry over. In cases where you go from incandescent to LED it's generally preferred to change the lens design"

Because previously, you said:

"The idea is that the OEM light uses bulbs and the aftermarket part will be LED, with a custom circuit board installed"

If you are copying the lens design like for like, I think you will have problems reproducing the optics and thus problems certifying the product. I must admit though that regulations for rear lighting is less stringent that for headlights. I hope you intend to properly check your legal status - I'm not sure it's that straightforward.

SRT Mike
06-30-2008, 01:53 PM
I don't understand why you now say:

"As for why trying to replicate, because we want to offer additional features beyond the OEM part. The lens design is applicable because the light I posted is LED so the lens design would carry over. In cases where you go from incandescent to LED it's generally preferred to change the lens design"

Because previously, you said:

"The idea is that the OEM light uses bulbs and the aftermarket part will be LED, with a custom circuit board installed"

If you are copying the lens design like for like, I think you will have problems reproducing the optics and thus problems certifying the product. I must admit though that regulations for rear lighting is less stringent that for headlights. I hope you intend to properly check your legal status - I'm not sure it's that straightforward.

I don't recall what pictures I posted (looks like they are gone now). I thought it was a bulb one. Anyway, we were interested in manufacturing lots of light assemblies. Some are OEM bulbs, some are OEM LED's.

As for the optics, it's not a problem. Many OEM brake lights don't use optics beyond what is built into the LED lens. There is no problem certifying the product - there isn't a certification process. The manufacturer claims adherence to DOT standards. People think that the DOT approves or denies designs - this is not the case. The DOT lays out the rules, and the manufacturer states they adhere or not. There are independent (not affiliated with DOT) labs that will determine if your product meets DOT or not, but the process isn't as daunting as most folks think.

I know the woman over at the NHTSA who is responsible for rear lighting and I talk to her frequently. I know when asked by another manufacturer of LED brake lights regarding the laws, they only wanted "rearward visibility in daylight at 300 feet". I think much of the time spent on design is based around using a smaller number of LED's for cost reasons, as well as making one part that meets all countries' standards (the Euros have different standards from DOT).

I'm not saying it's easy or anything, just that it's not even nearly as difficult and exacting as most think (and I used to think, prior to putting some stuff through DOT certification).

5axisguy
07-04-2008, 02:00 PM
I think it's my turn to chime in here, I have been in the moldmaking business for 22 plus years now and I will tell you there is no way to make a mold for that part for the 9500$ that is previously mentioned, unless of course its a prototype and a very crude one at that. I see ribs, possible slides, possible lifters, some sort of digitizing for the model, design, material, machining, edm, a good A-1 diamond lens quality mold finish, that would take at least 20 hours per core and cavity alone. Unless you find a mold shop working for like 15$ bucks and hour, your going to spend 4 or 5 times that 9500$ that was thrown out there. And yes I would think some legal issues would apply taking someones elses part and replicating it.

Mike Stevenson
07-04-2008, 03:03 PM
Word 5axisguy.

bgraham111
07-20-2008, 07:40 AM
I used to work for a company that did automotive lighting. Good luck, you are going to need it.

The LED lights won't come from the same point as a filament bulb, so all your lens surfaces will be off and will need to be re adjusted.

Surface finish is extremely important. Lens are very sensitive to surface finish.

The gate location and ejector locations are very important. Even if they are not in the surface that the light shines through, they can affect part quality (add light distortion.)

The only thing I'm not sure about is the legal requirements for after market parts. Maybe they don't have to meet the same requirements.