Messages

This is true, I always had plans to do what you did and write a program to hand pick files, never did. And I have never used the BOM extarction util because frankly its crap.
I quite like the halfway house, stick it in the machine to place high quantity things that were loaded anyway, manual the rest.

It's definitely worth it - sometimes even for 1-offs if there are a lot of common passives - for these jellybean parts, which usually live on the machine, I can go from the PCB file to having a JOB file ready in less time than it takes the RV PC to boot up, and the only setup needed is the PCB position so it can be placing in a few minutes.

..well that all depends on how many (different) parts there are, and how streamlined your programming process is....

Yeah I thought so too, but I suppose if that's all the cash you have, where else could you pay that little for a useable, small footprint P&P for prototyping & small runs if you don't mind using passive feeders?

Not my listing.
NB this is RV1 not RV1S - not sure of the difference - only info I have is that it has more limited vision and can't do 0603s

http://cgi.ebay.co.uk/Versatronics-RV1-Pick-and-Place-/160547973137?pt=UK_BOI_Electrical_Components_Supplies_ET&hash=item256167a811

This may be worth a look - haven't checked dimensions but readily available and cheap
http://www.sparkfun.com/products/10398

Depends what you want to do. The TFR has all the information on the design, from which the job file is generated, however as I understand it, things like feeder/board setups and how far through a placement it got are added to the job file during the placement process.
So if the purpose of the backup is to run the same job again in future, with  the same feeder setup, the job file may be useful as it should save some setup time. If however a new run of the job will be set up from scratch next time, the job file is less useful, and in fact generating a new one may be better to clear out old settings.   
Neither file is very big so no harm in saving both.

BTW the settings in the file above don't work if Windows display settings are  in 'large fonts' mode, however RV Place's screen redraw issues seem subjectively worse in this mode anyway so probably not an issue..!

The offsets between fid camera and nozzle do seem to increase in the feeder areas - I don't know if the grid calibration covers these areas, so this may at least partly explain it. You can at least get fairly close though.
If you've not already done so, check out my Joystick hack. This makes setup a lot quicker as you can keep your eye on the nozzle. A nice bright light helps - at some point I intend to add some white LEDs around the head to make it easier to see the nozzle.
Another thing to be aware of is that if the position is slightly off, the software adjusts the pick position based on the camera resault (unless the "lock feeders" option is set in the CDF), so a less-than-perfect position should progressively correct itself.

Yes - just click 'camera'  in the move-button dialog. Bear in mind that PCB refs don't need to be super-accurate as fid correction will correct any error.

The important file is RVDATA.DAT which contains machine-specific calibration information, so the only way to create a new one if the file is lost is with a full recalibration. 
There are also backup versions of the file with .bak and .sav extensions - I don't recall under what circumstances these are saved - looking at the timestamps may give a clue.

If  RVDATA.DAT is there :

Maybe the RTC battery has also messed up some other BIOS info, or something is getting confused by a strange time/date - this sounds like some sort of OS/PC issue causing the software not to be able to see the file. Check that the RVDATA.DAT file doesn't have a wierd timestamp caused by the RTC issue. Also check the file is readable - it's plain text.

Could it be that it was set up with the DAT folder  on a non-default drive, and the CMOS setup has lost some drive info making that drive inaccessible?
The file rvplace.cfg in the software install folder specifies where it thinks the folder structure should be. 
The only other small possibility  I can think of is that the CMOS settings related to things like the controller cards (e.g PCI configuration) have changed, and the error is actually somewhere other than the .DAT file.

I find 4mm wide double-sided tape (ebay) very useful for joining cover tapes and sticking cover tape to pickup reels.
I've found Farnell's re-reel service brass splices don't tend to travel well through RV feeders, and the feeder design can make clearing a jam a PITA to do without disturbing other lanes.

For leaded paste this stuff seems OK & pretty cheap, although a bit thick for use on plastic stencils.

The less-than obvious issue is that Automatic has three possible outcomes depending on size :  OTF, Fixed or Fixed+jiggle, and the lack of the third option in the menu implies that it should do it automatically.
Last time I had this it was with narrow SO16s, but a pre-rotate was just enough to fix it.
I've got into the habit of using Force Fixed as the first step when I hit imaging issues, and didn't realise this was what was stopping it doing the jiggly thing! Works fine now.

Just been having major problems reliably imaging these MOSFETs

Problem was it was not consistently including or excluding one or both of the small tabs at the side of the large pad, causing random small rotations as the tabs are outside the outline of the 8 main pads.
I spent quite a while trying different combinations of threshold & blob size with no joy.
Then I figured  the answer was to image the outline, with the aid of some tipp-ex on a spare nizzle to create a white background

Worked a treat - probably needs something a little more resilient than Tippex to stop it scraping off in the tool holder, but this was a 1-off build so it just about lasted.

Elektor unit looks quite good, although could do with being a little bigger. I've certainly seen less good ones for more money...
Video of the Elektor frame
Product page

Just wondering if there may be some dependance on Windows version or screen settings, as it's such a frequent occurrence that I'm surprised it hadn't been noticed & fixed at the time.

Maybe the question should be whether there is anyone that hasn't seen this...