LPKF PCB Prototyping

[Slow Poke]

I'm talking from experienece, as i previously owned a high end LPKF S62 Mill.     $30,000 of high end german engineering.

Was that a UV, possibly excimer, laser?  If so should be exquisitely good, able to etch microscopic traces, yes?

[mrpackethead]

no its a PCB milling machine.  100k spindle, great hardware, terrible software, and even worse support.   

[calvinlove0]

I just wanted to weigh in on this. I work for a university in Canada and we took a chance on LPKF. I completely agree with everything I've read in these forums - great hardware, glitchy software and awful support. Our S104 hasn't worked properly since we got it and its been completely out of commission for 5 months.

When we got the machine the USB cables needed to be replaced. They insisted I buy gold plated ones.

After replacing the USB cables the machine would hang up on tool check and the software would crash. As per the support instructions I pulled out the tool check dowel and cleaned it. Afterword it was too loose so I had to find the flow valve buried in the back of the machine and adjust. Took most of the day to calibrate the pressure on this one.

Finally we were milling boards...but then we had camera connectivity issues when running the solder paste dispenser.
After about a month of use, the S104 would just stop picking up tools altogether. They asked to send it back for repair, because we through out the crate that came with it we had to buy another for 400.00. Originally they said we would have to pay for one-way shipping then agreed to pay both ways after we hounded them.

They claimed they had solved the issue and sent it back a month later. Guess what...we're still having the same issues. They're now trying to blame it on the PC we're using. I've tried multiple PCs (all of which exceed the minimum spec listed in their manual).

I now have industry partners, researchers, students and instructors calling my constantly wanting to use this machine. For 30k, I really expected more from LPKF.

[KE5FX]

One useful figure of merit for USB cables is the shield resistance from one end to the other.  It's supposed to be less than 0.6 ohms shell to shell.  Yeah, right.  Their experience reflects mine as well, except that unlike them I have seen no consistent advantage with the more expensive cables.  Maybe 10% of the cables I've tested meet the spec, and the "gold" plated ones are no different than the rest.

It's very possible that the cable is an issue, especially in a noisy RF environment.  But you have to measure them by hand in order to find a good one. 

[G0HZU]

I've only ever used the older LPKF machines like the C60. The one at work is about 20 years old now and is still working fine. The older LPKF and T-Tech machines used RS-232 rather than USB. The even older T-Tech machines used an internal AT card in a PC and used a large parallel interface direct to the milling bed. This parallel interface is not the same as the LPT parallel port on a PC.

The later RS-232 T-Tech machines used RS-232 from the PC to an intermediate control box which then had the large parallel interface to the milling bed.

In all cases, the advice from the manufacturer was to use a dedicated PC to control the milling machine. This means, no networking, no other SW installed apart from the main OS and the BIOS had to be set up as per the manufacturer's recommendations. There will be a steady stream of control data from the PC and the PC must not have any distractions from networking or power management tasks running in the background. Otherwise, I think the system can go out of synch and trash the board being milled.

The very early T-Tech machines had to have the PC's internal cache disabled to prevent problems. The later machines were much more tolerant and stable. I've got a couple of old T-Tech 7000S models here and these use RS-232 to interface to the PC. I'd imagine that it will be much more challenging to do all this via USB on a modern PC. I'd recommend using a barebones PC that is dedicated to the milling machine and nothing else. Otherwise, the PC could divert itself to do some background housekeeping tasks and this could cause gaps in the stream of USB control data from the PC. This could upset the system.

Here at home, I use an old Win 98SE laptop to control my milling machine via RS-232 and it works flawlessly as long as I service the machine regularly. The moving parts need to be cleaned and oiled/greased several times a year. I've had my main T-Tech 7000S for over 20 years now.

[tomgat]

They claimed they had solved the issue and sent it back a month later. Guess what...we're still having the same issues. They're now trying to blame it on the PC we're using. I've tried multiple PCs (all of which exceed the minimum spec listed in their manual).

I now have industry partners, researchers, students and instructors calling my constantly wanting to use this machine. For 30k, I really expected more from LPKF.

My suggestion is to not over think the issue.  Essentially all this machine does is have a generic USB camera attached.  Nothing special.  The S103 uses an old microsoft vx-800 life cam camera.  Which camera does the S104 use? 

Your issue most likely is an intermittent cabling issue which, in fairness, would be tough for LPKF to diagnose.  My suggestion is to break down what you know and dont know.  If you let the machine sit for a while, does it lose camera connection, or does it only loose the camera connection when moving?  I would suggest unmounting the camera and looking at its cabling.  It might have a bad connection to its camera board.  If the cabling looks good, then you might just want to buy a new camera (I bought a generic VX-800 microsoft camera for $20 on ebay, and replaced mine once before)

In short, LPKF support (Tim is the most knowledgable BTW) are smart, but they wont know to the level you likely need.  Your the EE.   Just my 2 cents.