Friends, stage 1 polishing has been completed, which makes the plastic mold 85% complete.
Polycarbonate dyes will also arrive tomorrow, so tomorrow is Factory Day! The temperature here is 36-37C and the factory has no air conditioning (it's being renovated). We will see our mold in action for the first time... and if we survive the heat, expect videos and plenty of pictures within 24 hours, documenting:
- the first DL2000 molded enclosure samples EVER
- different colors and intensities to compare
- BRASS INSERTS which will be overmolded to allow you to open and close the receiver an unlimited number of times (they prevent damage to the walls of the screw holes)
Because the mold will have to undergo another round of polishing, the plastic samples we'll get tomorrow might NOT be as clear as glass - no idea what to expect honestly.
We will own the first official DreamLover 2000 molded plastic enclosures on Friday!
Jason's off to Beijing for his Confucianism lesson (we still think he's got a girlfriend). His assistant will help translate foreigner Mandarin to native Cantonese and his workers will load the mold into the injection machine and shoot out some parts. What an emotion. It seems like yesterday that we were looking for non waterproof square boxes on the net and making holes with a hand drill.
And on Friday a million dollar injection molding machine with our ton and a half custom carved and polished mold loaded inside will be pumping out Male Management paraphernalia for the first time ever on planet earth.
That's not all. We'll be able to test 3 different dyes which will be mixed with the polycarbonate plastic. These will give the plastic a slight tint. The limit of 3 comes from the fact that the mold needs to be washed after each tint is loaded and because it's heavier than an automobile it's something that takes time to accomplish.
These dyes are only useful if we do not spray paint the receiver.
We've chosen three colors:
- black: because it should give the product the nice outline of the spray painted parts while retaining transparency
- white: because it should diffuse the LED light enough to blend the colors well and hopefully spread the color throughout the enclosure
- purple: because it's just an awesome color and it's part of our "corporate" color scheme
How tints will affect the LED remains to be seen. The tint has to be strong enough to hopefully make the product look more finished and attractive, but light enough to allow the LED to shine through without blocking out some colors. It should also be light enough for you to be able to see if any water went inside.
These tints should work very well with the 3D shape of the receiver; depth is different in different areas, and this produces a nice effect near edges (flat areas are a lot lighter, thicker areas have more intense color).
We wish we could try out all pantone colors but it would cost us a million bucks :) So for now let us rejoice about this monumental step forward and enjoy whatever colors we're allowed to test.
A few more curiosities. We've asked Jason about putting TWO color dyes in the mold (or chunks of one color), to create some unique non-repeatable color streaks and patterns. He said this would be hard to control. We will take a chance on Friday to try to add a dye without washing the mold and see what happens. Honestly we've got no clue whatsoever how these machines are operated. It may not be as easy as throwing powder in like in a washing machine :)
Also, the molded parts on Friday will still not be 100% final; the mold will undergo yet another polishing phase at another location to make it mirror-like and perfectly smooth. It should still already be much smoother than our machined parts though.
We still have the option of adding texture after that, but it's a one way thing (the mold is carved with some chemicals). To undo the texture the mold needs to be re-polished and that will add 0.1 mm thickness or so.
As for the remote and DL2K-LINK enclosure, you probably already know that those will be sprayed black. Unless the purple remote with the black overlays looks mighty cool which it should.
A very hard day of soldering and the boards (3 sets of receiver and transmitter) are ready for lab testing.
Total headache when one of the RF modules was found to be defective after it was soldered!
This castellated 12 contact module is very tricky to desolder. Two soldering irons (and two hands) were used for lack of a better tool. Even with a heat gun it is probably very tricky to get it off because there are many 1x0.5mm SMD components on it and these will come off a lot quicker than the pads on the sides...
Fortunately the chip was desoldered and replaced successfully with the two irons technique and now both the two new receiver prototypes work giving us a total of 3 working DL2000 units.
Well... we can't afford NOT to machine assemble these. Putting them together requires prolonged, extreme concentration - it is exhausting work! On the positive side the accelerometer can now be placed without any problem because we've gotten enough practice with the 20 pin 3x3mm leadless package (battery charger IC) that now this 8 pin one seems "easy".
This product is so impossibly miniaturized that even a 1mm error in the length of a through hole component (LED, pulse capacitor) leads makes it impossible to close the enclosure! This time we screwed up the length of one LED and had to cut a piece off :) In the future the position of the LED must be watched more carefully.
Oh, we need to get full color LEDs now. Turns out the supplier never shipped our paid order. Too many things to worry about. Lastly, because we ran out of the right part and used a replacement, one of the LEDs used has inverted green and blue pins (hence the color difference!)
Hard day... soldering this product by hand is really exhausting.
Due to lack of time we will quit updating the R&D page and instead use blog entries tagged 'release timeline'.
The present status as of today:
- we're waiting for the mold to be Electric Discharge Machined and polished. In 4 days we should be able to go to the plastic factory and get the very first DL2000 molded plastic parts. Exciting!!!
- if the parts are ok, there will be an additional polishing stage which is to be carried out at another location. Mold needs to be transported on a truck. Should take 1 week
- presently the bank's giving us problems with documents and the incorporation company lady is not being very polite with us. They've increased our price and handled us with a "take it or leave it" attitude when the process is already half way through and we can't even switch to a different vendor! We had to agree to the price increase, and they are now working on it and not really answering emails. We hope that in 2 weeks they can get this account open! Our first time, beyond our control, fingers crossed.
- today we're assembling 3 sets of the boards (remotes done already, receivers 2 to go); these can be used for EMC testing as soon as we get the molded parts and have time to go to Mr Tang's rubber factory and produce one arousal inducer with the capacitor inside (the capacitor's used to reduce emissions without resorting to the ferrite bead which would add thickness - will it work???). We plan to do the lab tests within 3 weeks, and hopefully we'll pass the first time because this is in our estimation the biggest waste of money in bringing a product to the market!
- we're working to get body bands (with some means of fastening them) and boxes. We're already receiving samples (trying to make the most out of living in China). Finalizing requirements and getting something nice created and getting a reasonable quote may take 3-4 weeks but these have low priority as the product is usable without them and they can be added later
- some paperwork is still missing for legally selling the DL2K-LINK! We're talking with our contacts to get this resolved and hope that within a month all will be in order.
- the boards have been tested to death. Yes there could be some improvements as far as moving components around for easier mounting etc. but at this point we're committing to our design. When we pass the lab EMC tests we'll be ready to produce a few hundred boards and start experimenting with automated assembly. We say experimenting because some parts will likely have to be hand mounted anyway, due to the fact their packaging is not suitable for pick and place machinery. Though the assembly house guarantees they can do it, we trust the manufacturer on this - they deal with lots of assembly houses in the west. PCB production won't take more than 10 days, usually 1 week. Assembly - we don't know, they said 1 week, anyway if it looks like this will delay us we'll just hand assemble boards for our first customers.
- we still have to reimplement a bunch of functions which were tested on older models and are not in the newest firmware. Accelerometer on/off switching, dog mode, timer mode, wireless reprogramming of the REMOTE (receiver already done), moving the reprogramming to the boot sector so the device can't get messed up when some packets are corrupted, learning remote IDs and so on. The PC control panel needs to be updated. The web interface needs to tie into our main database with the user data. This is just programming work. It will be finished by the time everything else is finished.
- the very last step is to figure out who is going to assemble the parts of this product and where (pcb inside the plastic, cables, grommets, etc), and go through the process of putting product on the shelves of the fulfillment houses which involves crossing national borders (customs) and trying to do this with as little tax overhead as possible. For Europe we even need to pay some company to hold our documents within the EU! Bah. Our first time - this shouldn't take long provided we can make a video that explains exactly how to mount the parts... hoping we're not underestimating this phase.
There haven't been new "to do" items for over a month so it's unlikely something will pop up that will delay us further. At this point it's just a matter of going through these necessary steps one by one. If we're lucky we may be trying to figure out customs clearance and how to receive funds within 30 days.
We'll keep posting our progress. Things are moving fast now.
