I ordered a TM240A (the larger machine) and the total cost (including shipping to Edmonton, Canada) was US $5071. The ship cost was $595, which was considerably higher than what I had expected, based on Ian's experience. But we went for it anyway.
I am indeed hoping to use this in a production environment, for small runs of 10 to 25 panels at a time. Most of my SMT boards are tiny, so each panel has anywhere from 6 to 12 boards. I'm now ordering paste stencils from the PCB supplier - they are supplying stainless-steel stencils for US $50. I did ask if they can drop that price any - for my most recent PCB order (late last week), they dropped the stencil cost down to US $45. I'm not going to push them any harder - I think that $45 is a reasonable price to pay for stainless.
I have heard that Seeed Studio will be offering a stencil service with their boards - I may have to check that out as well.
I'm hoping that my machine arrives late this coming week or early next. I'm currently looking at using one of my text editors (TSE or UltraEdit) to see if I can automate converting the PnP placement data from my CAD package into the format that the TM240A wants to see. But its been literally * YEARS * since I did any scripting with the macro language - I used to be pretty good at it but I'll be starting from scratch again. I'll most likely start with TSE (from http://www.semware.com
) - I still use a heavily-customized version of Qedit (very, very early version of TSE - now know as TSE Junior) as my daily editor and I recall that it was fairly easy to make the macro language do what I wanted.
I still think that this machine will make a fine production machine. Do recall that all of the early commercial PnP machines were based on dead-reckoning and did not have vision assistance. Quite frankly, I think that the low cost PnP machines that do have vision are using the vision system to compensate for poor mechanical accuracy. These machines without vision will have to be darned accurate - they have no choice if they are to be usable. Certainly, the specifications do state that these machines are darned accurate - far more so than the early Zevatech machine that I worked on 15 or 20 years ago. I never did see the Zevatech machine operating but its owners certainly made a lot of money from that machine - they were a tiny contract manufacturer here in Edmonton and that machine was the central part of their business. But I flat out guarantee that it was not as accurate as the specifications of this machine.
Whether or not this machine meets its specifications is something that I am quite looking forward to finding out. <Grin>.