[Community] FirePick Delta desktop electronics manufacturing
Dr. H. Nikolaus Schaller
hns at goldelico.com
Wed Jul 30 22:34:24 CEST 2014
Am 30.07.2014 um 20:59 schrieb Robert 'Bobby' Zenz:
> They are aiming at a price of $ 300.- for one machine, see their store
> Might be wishful dreaming, but if they pull if of for $ 3000.- it's
> still a win.
For that money you can run 10 projects like the Neo900 at some external EMS
(only counting pick&place machine operation - nothing else which contributes
to production cost).
Sometimes it is cheaper not to buy equipment, but to rent it if needed...
I.e. let somebody else do the investment and spend a lot of money for
a machine that otherwise collects dust 360 out of 365 days per year :)
The same holds for PCB production (still people use laser printers
to try to make etch resist out of it and do etching themselves where we
can get double layer samples at ~60 EUR).
It is nice to learn how such production processes work, but it is not for
running a project like a GTA04 or Neo900. There you need professional
tools to get professional results.
Just my PoV.
> On Wed, 30 Jul 2014 19:26:08 +0200
> "Dr. H. Nikolaus Schaller" <hns at goldelico.com> wrote:
>> Am 30.07.2014 um 14:59 schrieb Ryan de Laplante (personal):
>>> This morning while reading Slashdot news I came across this project:
>> It is a nice idea.
>>> I wonder if it could be helpful to reduce assembly costs of Neo900?
>> No. Not really.
>> Those commercial machines are more expensive for good reasons...
>> They have:
>> a) a much higher mechanical precision
>> b) automatic feeders for components coming as cut tape - this is
>> important to get them to speed c) they are faster
>> But, we do not have to buy them. We just rent them for let's say two
>> days in which they run round the clock to populate 200 boards. This
>> is very affordable.
>> If such a machine costs 100000 USD and runs for 10 years on 220
>> working days per year, it costs them just 45 US per day. So they will
>> charge us let's say 300-500 USD for renting (and maintaining) the
>> machine that is to be replaced by a device like a firepick. So buying
>> such a firepick doesn't look attractive any more :(
>> All other costs are not reduced (don't forget to buy components and
>> PCBs and solder paste). This pick&place step is not the most
>> expensive part of building a device or production cost.
>> Other parts of production cost are manpower to purchase components,
>> keep stock, build pick&place programs etc and set up the machine and
>> keep it running (i.e. trouble-shoot, supply solder paste, new reels
>> of components coming on tape or replacing trays).
>> Next, it is much more important to manage the solder paste printing
>> process and the soldering process itself (which is not improved by
>> this machine because these steps come before or afterwards).
>> Finally, I have done today manually what such a machine could do. I
>> have done it using a good stereo microscope and tweezers... To build
>> new Pyra-handheld prototypes. Yes, that is a slow process and you
>> need a calm hand, but it is doable. But of course not for 200
>> units... That would drive one crazy :)
>> So in general such a machine is just an alternative for one of
>> several production steps that are needed.
>> Which means that all others still have to be done externally. So we
>> simply ask some EMS who can do all steps for a quotation and it is
>> very unlikely that they will replace their good pick&place machine by
>> something else...
>> Community mailing list
>> Community at openphoenux.org
> Community mailing list
> Community at openphoenux.org
More information about the Community