Author Topic: LPKF  (Read 25480 times)

0 Members and 1 Guest are viewing this topic.

Offline s103_protomatTopic starter

  • Newbie
  • Posts: 6
LPKF
« on: May 27, 2014, 09:26:15 pm »
I have been working with the below equipment from LPKF for over 6 months now and consider myself to be quite knowledgeable with their systems. I use this equipment essentially every day and have a serious love/hate relationship with it. The equipment is quite capable of making dense multilayer boards in a very short amount of time but the amount of time it takes to learn each and every "quirk" in the system makes it almost worthless.

LPKF has a support staff to help with minor issues such as tool ramp calibration and tool width adjustments but when asked a more complicated question such as "why does copper leak into my internal planes and create shorts?" you might get an answer like "I have never seen this issue before and i have been at LPKF for years" :palm:. Alternatively they will have you make unrelated adjustments until you get so frustrated that it becomes more time effective to try new things on your own |O. In general I only call LPKF these days if I feel like punching myself in the face.

It is for this reason I will be posting every issue I have come across in my 6+ months using the equipment. Generally speaking not one of the below machines/processes works perfectly. They all have underlying engineering issues and plenty of software bugs. I am not sure how long it will take me to get to all of this but I promise to get it posted as quickly as possible. If you would like some expedited help feel free to post on this thread and I will get to it ASAP.  :-+



LPKF Equipment List:

CircuitPro 2.1                          (7 out of 10)
   -Gerber files from Altium
   -one to six layer imports
   -Windows 7 (64 bit)

S-103 milling machine             (4 out of 10)
   -Vacuum table
   -auto tool change
   -Floating head
   -Fiducial camera
   -Paste dispense

MultiPress                                (6 out of 10)
   -Hydraulic pump

MiniContac                               (5 out of 10)

ProMask/ProLegend :--            (Negative 10 out of 10)
   -Laser printer (HP M1212nf MFP) (1200x1200DPI)

Circulating Oven                       (7 out of 10)
   -Cadco, Roberta


 

Offline mamalala

  • Supporter
  • ****
  • Posts: 777
  • Country: de
Re: LPKF
« Reply #1 on: May 27, 2014, 10:38:56 pm »
Inner layers? With milled PCB's? Seriously? How is that even supposed to somewhat reliably work?

I never understood the frenzy about those PCB milling machines anyway. They are slow, they are extremely limited when it comes to trace and restring thicknes. They produce a lot of dirt. Due to the fact that the copper is mechanically removed, you get a huge impact on the electrical characteristics of the board compared to regular etching. Plus, the tools are quite expensive and don't last that long either, if you use FR4 material.

But really, can you explain to me how the process for a multilayer board is supposed to work? And i mean the reliable electrical contacting of the layers through the via's. And the stacking of the boards. Unless it also uses regular through-plating, that is. In which case you still would get better results by going the industry proven photo route.

Greetings,

Chris
 

Offline M4trix

  • Frequent Contributor
  • **
  • Posts: 307
  • Country: hr
Re: LPKF
« Reply #2 on: May 27, 2014, 10:50:14 pm »

ProMask/ProLegend :--            (Negative 10 out of 10)
   -Laser printer (HP M1212nf MFP) (1200x1200DPI)

Doesn't look too bad to me.  :-//

 

Offline mamalala

  • Supporter
  • ****
  • Posts: 777
  • Country: de
Re: LPKF
« Reply #3 on: May 27, 2014, 10:56:58 pm »

ProMask/ProLegend :--            (Negative 10 out of 10)
   -Laser printer (HP M1212nf MFP) (1200x1200DPI)

Doesn't look too bad to me.  :-//



The problem is that you rarely get a print from a laser that dimensionally matches 1:1 to whatever is produced on a CNC. Due to the fuser unit, and the heat applied in there, you will pretty much always get a distorted result. And even if the outer reactangle matches, you can still have distortions inside. Not to mention that home printers are rarely 1:1 when it comes to actual meassurements. What you draw up as 10mm might well end up 9.5 or 10.5mm in the printout, but almost never 10mm.

For coarse THT stuff this may be OK, but as soon as you go SMD you are virtually lost. The only way to get a repeatable and in-spec (measurement wise) result is to have a litho produced. But then, you can put the layout on the litho as well, and do your PCB the photo way all along.

Greetings,

Chris

Edit: And what with all the paint-on stuff? Why not use simple, easy to use resist here, like Dynamask?
« Last Edit: May 27, 2014, 10:58:54 pm by mamalala »
 

Offline M4trix

  • Frequent Contributor
  • **
  • Posts: 307
  • Country: hr
Re: LPKF
« Reply #4 on: May 27, 2014, 11:15:54 pm »

Edit: And what with all the paint-on stuff? Why not use simple, easy to use resist here, like Dynamask?

Yes, Dynamask is an excellent choice.  :) 
 

Offline mamalala

  • Supporter
  • ****
  • Posts: 777
  • Country: de
Re: LPKF
« Reply #5 on: May 27, 2014, 11:21:58 pm »

Edit: And what with all the paint-on stuff? Why not use simple, easy to use resist here, like Dynamask?

Yes, Dynamask is an excellent choice.  :)

Yes. It gives you repeatable results, is easy to use, and best of all, you only use what you need (i.e. cut it to size before applying). With that paint-on stuff, you end up with a lot of unused stuff that you throw away later, need to clean up all the utensils quickly, etc. In an industrial setting it is useful. But then it is usually something like a dip-in bath or the likes, and since they are up producing most of the time and in large scales, there is little waste.

Greetings,

Chris
 

Offline mamalala

  • Supporter
  • ****
  • Posts: 777
  • Country: de
Re: LPKF
« Reply #6 on: May 27, 2014, 11:31:23 pm »
Just to add:

I mean, if someone goes to such lengths to produce prototypes, that is, using proper soldermask, silk-screen, multilayer, etc... Why not dojng it right to begin with? If you think your self-made prototypes are that important, or if you have such a large volume of one-off's, why use crap like a PCB mill? Use proper litho/film, expose the PCB, develop and etch. That will be faster than this milling crap, and with less dirt. If you have problems producing suitable stuff to expose the PCB's (i.e. like simple lase printers), go and grab an old Linotronic + RIP, and a developing-machine, and be done with it (similar old units available from Agfa and many others, you can get them for cheap used). Much more resolution than what you could get with milling any day.

Once you have that photo stuff for the PCB itself, re-use it for the soldermask. The only extra machine you need is a lmainator. And while at it, to produce a silk-screen you can use the same photo stuff again. Just check up these t-shirt printing things. It's a screen mesh where you apply photosensitive paint, let it dry, expose & develop, and then print thhrough. All you need is a screen with a finer mesh if you go for PCB legend stuff.

But really, spending thousands on these PCB milling machines, and then wasting lots and lots of time, only to have a 50/50 chance to get something useful ... really? All the work, dirt, consumables, etc., just for that?

And that, mind you, is only if you really need/want to do it on your own anyways.

Greetings,

Chris
 

Offline MasterBlaster

  • Newbie
  • Posts: 6
Re: LPKF
« Reply #7 on: May 28, 2014, 01:49:44 am »
I'd like to hear more about your work-arounds although I won't be buying one anytime soon. I thought about buying an older LPKF PCB mill sometime back, but the price was still too high. Etching isn't that easy if you ask me and waiting 2-3 weeks for every iteration from China is a pain in the a** sometimes.
« Last Edit: May 28, 2014, 01:54:01 am by MasterBlaster »
 

Offline mamalala

  • Supporter
  • ****
  • Posts: 777
  • Country: de
Re: LPKF
« Reply #8 on: May 28, 2014, 02:39:25 am »
Etching isn't that easy if you ask me ...

If you think that etching isn't easy, just wait until you get your hands on one of these PCB mills. Better stock up on loads of bare PCB's, as well as the milling cutters needed, which that machine will eat like nothing else...

Greetings,

Chris
 

Offline MasterBlaster

  • Newbie
  • Posts: 6
Re: LPKF
« Reply #9 on: May 28, 2014, 03:17:17 am »
Etching isn't that easy if you ask me ...

If you think that etching isn't easy, just wait until you get your hands on one of these PCB mills. Better stock up on loads of bare PCB's, as well as the milling cutters needed, which that machine will eat like nothing else...

Greetings,

Chris

Actually, I'm waiting around for some open hardware laser technology. :) LPKF has something but it's outrageously expensive of course...
http://www.lpkf.com/products/rapid-pcb-prototyping/laser-circuit-structuring/laser-structuring-printed-circuit-boards.htm

I wonder about the fumes though... vaporizing copper??? Eh....
« Last Edit: May 28, 2014, 03:20:05 am by MasterBlaster »
 

Offline mamalala

  • Supporter
  • ****
  • Posts: 777
  • Country: de
Re: LPKF
« Reply #10 on: May 28, 2014, 03:26:23 am »
Actually, I'm waiting around for some open hardware laser technology. :) LPKF has something but it's outrageously expensive of course...
http://www.lpkf.com/products/rapid-pcb-prototyping/laser-circuit-structuring/laser-structuring-printed-circuit-boards.htm

I wonder about the fumes though... vaporizing copper??? Eh....

Still doesn't make much sense. There have been lots of people who abused a pen-plotter equipped with a UV laser to expose photosensitive PCB material, for example. The issue is time. It simply takes way too long. Same goes for even simpler stuff like toner-transfer. In the time needed to print the layout, transfer it to the PCB, rub off the paper ... I have exposed & developed a handful of boards already.

Oh, and with a laser "etching" away the copper you get, again, the problem of first adjusting the machine properly, or you will burn lot's of PCB material. Plus, what about the power requirements?

I simply fail to see the benefit in any of these techniques. "no chemicals" is misleading. Where does all the evaporated copper go?

Greetings,

Chris
 

Offline PlainName

  • Super Contributor
  • ***
  • Posts: 6769
  • Country: va
Re: LPKF
« Reply #11 on: May 28, 2014, 10:37:15 am »
Quote
fail to see the benefit in any of these techniques

You says that 'no chemicals' is misleading,  but it's a significant factor: my mill sits next to my PC and disposal of copper is via a vacuum hose (hooked to a car vacuum cleaner). A domestic vacuum cleaner deals with anything not sucked up.

OTOH, when I did etching I had a big tank of corrosive stuff that would leave splashes no matter how careful I was, I needed plumbed-in water for cleaning, and a separate waste system for disposal of the cleaning water. That alone made it unattractive: I wound up with a big yellow bucket on the floor, the etch system on the counter top, a hose to the water tap, and waste outlet from the tank to the bucket. Making sure I finished before the bucket got full I still needed to cart the stuff out to a safe area of the garden to empty it. I sure ain't going to do that whils the missus makes dinner, but I've made milled boards at 1 or 2am as an alternative to Minesweeper just before bed.

Plus, the mill accurately and quickly puts the holes in the board and shapes it to boot. What's not to like?

Quote
the milling cutters needed, which that machine will eat like nothing else

You can extend the life significantly with lubrication. I now spread a little 3-in-1 over the PCB before starting, and add drops if the milled glass soaks it up too much. There is not enough to spit anywhere, and as a bonus it stops all the dust so you don't need a vacuum or whatever. Tools last much longer and the edges of the milled tracks are much cleaner.

Yes, I am aware that 3-in-1 could be classed as 'chemicals' :)
 

Offline mamalala

  • Supporter
  • ****
  • Posts: 777
  • Country: de
Re: LPKF
« Reply #12 on: May 28, 2014, 06:51:16 pm »
You says that 'no chemicals' is misleading,  but it's a significant factor: my mill sits next to my PC and disposal of copper is via a vacuum hose (hooked to a car vacuum cleaner). A domestic vacuum cleaner deals with anything not sucked up.

Well, it all depends, i guess. There are bubble etching tanks available for cheap, where you can leave the FeCl3 in. All you do it put in the board, turn on the air-pump and heater, wait for it to be done, and then take it out and turn off. It's also possible to extract the copper from the FeCl3 if wanted, which makes it more or less disposable through the sewer. After all, they are throwing a lot of that stuff into the waste water at sewage treatment plants... A water bath used to rinse the board after etching can also be re-used a lot of times before it needs changing the water.

Yes, the drilling is nice, no doubt about that. Personally, i would go for classic photo method, FeCl3, and as a bonus drilling through a CNC. Guess it all depends on how often one needs to make their own boards. However, i was mainly responding to the OP, where it seems that rather complicated stuff is done, like multi-layer boards with inner layers, stop-mask, silkscreen. So there already is a lot of other stuff involved, in which case (as i said) it makes no sense to me to use a PCB mill for the PCB's. The only extra chemical in all this would be the etchant, since developer and stripper is already needed, plus whatever is used for the contacting to the inner layers. This is obviously different from making simple 1 or 2 sided PCB's with no extras.

Greetings,

Chris
 

Offline PlainName

  • Super Contributor
  • ***
  • Posts: 6769
  • Country: va
Re: LPKF
« Reply #13 on: May 28, 2014, 10:06:31 pm »
Quote
There are bubble etching tanks available

Yes, I started with one of those, and then graduated to the Mega combined etch and wash. The area around each gets spattered no matter how careful one is (it is, after all, a tank blowing bubbles of etchant).

Quote
possible to extract the copper from the FeCl3

Most things are possible :) But it's another messy thing to deal with. Also, our local sewage is a sceptic tank, and it's not in our interest to let any of this stuff get down there.

Quote
The only extra chemical in all this would be the etchant, since developer and stripper is already needed,

I looked at it the other way: since I'm going to be using CNC to cut out the board and put holes in, it's not more aggro to mill the tracks too.

But this is based on my requirements - if I were intending to run off several of a board, or have quite complex ones, I would probably do it differently. In fact, I would send them off to itead or similar!

But that's because I'm no longer set up for etching. Sometimes I think a job would benefit from etching rather than milling, and if I were set up for it I'd do that, but since I'm not it's just too much hassle to get going again. The chap may be doing his multilayer boards through milling because that's what he's set up for. Someone else who normally etches would see if differently.

Another thought: I would think that accurate hole drilling is pretty important on multilayer since inner layers slightly offset would make the holes effectively smaller. I haven't done it, but I would think that drilling holes in the assembled stack might be tricky because you need to align the stacks, and using the holes to do that would seem to be a no-brainer to me. So if you're going to use CNC for those you might as well go the whole hog. Takes longer, but once you've set it in motion you can go off and do other things until it's finished.

 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3007
  • Country: gb
Re: LPKF
« Reply #14 on: May 28, 2014, 10:50:18 pm »
At my place of work we have been using an LPKF Proto 60 for about 11 years now. It's had a few service callouts over the years (so it hasn't been totally reliable)  but it has been a very useful tool :)

I also have two T-Tech 7000S milling machines here at home and I bought the first one some time around 2001. Anyone who can't appreciate how useful these things are probably hasn't used one to its full potential. You can mill front panels, engrave things, make laminated tools as well as making PCBs. I even milled a flexi PCB on mine.

However, to get the most time/cost effective benefit compared to other PCB manufacturing methods you need to be milling small boards on exotic PCB dielectrics. Milling a PCB is mainly a serial process because it's like drawing a picture.
It's possible to mill a simple PCB (eg a distributed element RF filter) in a few minutes and if it tests out 'wrong' then a new version can be milled in a few minutes.

But where it gets very inefficient is if you try and mill something like a large PCB with a CPU and associated address bus and data bus going to loads of other devices on a two layer PCB. It can literally take hours to mill the board because there are so many fine tracks to mill.

By contrast, etching a PCB is like a parallel process because everything gets etched at the same time. So a big and complicated PCB etches in roughly the same time as a small simple PCB.

The milling machine easily 'wins' for small RF PCBs because it can mill a simple artwork and drill the holes and mill out highly accurate cutouts in the PCB as well as milling a complicated PCB outline and do it all in a few minutes. Preparing the artwork plots doing the resist, getting out the chemicals and etching and drilling and sawing the PCB outline and cutouts by hand is not much fun and will take a lot longer...
« Last Edit: May 28, 2014, 11:32:05 pm by G0HZU »
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26682
  • Country: nl
    • NCT Developments
Re: LPKF
« Reply #15 on: May 29, 2014, 04:45:17 pm »

ProMask/ProLegend :--            (Negative 10 out of 10)
   -Laser printer (HP M1212nf MFP) (1200x1200DPI)

Doesn't look too bad to me.  :-//



The problem is that you rarely get a print from a laser that dimensionally matches 1:1 to whatever is produced on a CNC. Due to the fuser unit, and the heat applied in there, you will pretty much always get a distorted result. And even if the outer reactangle matches, you can still have distortions inside. Not to mention that home printers are rarely 1:1 when it comes to actual meassurements. What you draw up as 10mm might well end up 9.5 or 10.5mm in the printout, but almost never 10mm.
That is a bit too strong and maybe true for cheap rubbish printers. HP laserprinters however are very accurate. I have been using HP laserprinters to print PCB layouts on tracing paper for over 15 years. Printing & etching footprints with a 0.5mm pitch is not a problem at all and the parts fit perfectly.

Then again Eurocircuits and Seeedstudio are so cheap nowadays that it hardly justifies making your own PCBs. I only etch my own PCBs when I need a prototype really fast.
« Last Edit: May 29, 2014, 04:48:45 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline mamalala

  • Supporter
  • ****
  • Posts: 777
  • Country: de
Re: LPKF
« Reply #16 on: May 29, 2014, 06:42:42 pm »
That is a bit too strong and maybe true for cheap rubbish printers. HP laserprinters however are very accurate. I have been using HP laserprinters to print PCB layouts on tracing paper for over 15 years. Printing & etching footprints with a 0.5mm pitch is not a problem at all and the parts fit perfectly.

Of course that is not a problem. But then, it wasn't the point i was making, either.

Take a doublse sided layout. Print it on a suitable (half-)transparent medium. Now see how well both sides match each other. Same goes for having a PCB drilled by a CNC, when the PCB was exposed/developed/etched with a transparency made in a laser printer. This is simply due to the way a laser printer works, that is, the fuser that melts the toner onto the paper/film/whatever. This will cause the media to distort.

Greetings,

Chris
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26682
  • Country: nl
    • NCT Developments
Re: LPKF
« Reply #17 on: May 30, 2014, 08:47:18 pm »
Again: I never had problems with that. In the past I have made lots of double sides boards with printouts on tracing paper. Mismatch between the printouts has never been an issue for me.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Mr Simpleton

  • Supporter
  • ****
  • Posts: 302
  • Country: se
  • Not the sharpest knife in the drawer
Re: LPKF
« Reply #18 on: June 03, 2014, 10:32:34 am »
I have the manual LPKF-39 which is the thing for fast prototyping RF boards... you can even draw the layout using pen and paper.  Now we do have two computer controlled LPKF which is fine but they do not offer automatic tool cange. Still great for single side PCB, and do work in a pinch for doubble sided. Still the through hole sytem we bought never produced anyting I could use, the unplated doubble side PCB can be a pain if using multi row connectors!

So we have pretty much reverted to order etched boards, with turn-around of a week, gives us nice solder stop masks and through hole plating. Still I find my self using the LPKF when needing one-off board a friday afternoon, or if my hobby project needs a board done :)

Add me to the love-hate relation with LPKF, but with my "39" its pure love  :-+
 

Offline VK5RC

  • Supporter
  • ****
  • Posts: 2672
  • Country: au
Re: LPKF
« Reply #19 on: June 04, 2014, 05:27:20 am »
G0HZU,  does it take long to learn how to program these mills? What program do you use at home? (for board layout and mill control),  Thanks Rob
Whoah! Watch where that landed we might need it later.
 

Offline Mr Simpleton

  • Supporter
  • ****
  • Posts: 302
  • Country: se
  • Not the sharpest knife in the drawer
Re: LPKF
« Reply #20 on: June 04, 2014, 03:07:55 pm »
If you are interested in LPKF, I may help out...
You do design in your normal CAD program, and export the "normal" photo plot in Gerber, and drill info as Excellon.
Make note on how your settings are wrt leading zeros and number of decimals.
Now open up Circuit Cam, which can import these formats and you do set up the numbers you previously noted, and make these default. If your program settings are correct wrt to planes naming, Circuit Cam will import and place the routing on correct planes, otherwise you do have to manually do so.

After verifying that all looks good and the holes lines up with the pads, next step is mill routing, and you can even select that twice the insulation distance is needed around pads. Then time for cutting out the board, and you even get tabs so the board will stay in place.
Most settings may need a tweak the first time you rund but becomes default after that.

Now with all the routing done, you export this, in a format which your LPKF mill control s/w undstands.
The milling s/w will promt you for selecting proper tools and if you need to flip the board over.

It may be a bit tricky to get all settings at first, but not too complicated.

Not sure how the T-Tech operates but should be pretty similar.

BTW saw a new machine offered, http://www.cirqoid.com/ nice price vs. LPKF, but have yet to try it out!
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3007
  • Country: gb
Re: LPKF
« Reply #21 on: June 04, 2014, 10:57:26 pm »
G0HZU,  does it take long to learn how to program these mills? What program do you use at home? (for board layout and mill control),  Thanks Rob
In the case of the old T-Tech 7000S you have to be careful to get the later version with the variable spindle speed. My first one has fixed  speed at 23,000rpm. This fixed speed version isn't supported by the newer/better Isopro program that controls these mills.

The only software that runs with the fixed speed version is an ancient and horrible DOS program. In desperation I ended up reverse engineering the isopro software to work out which part of the program expected to find a variable speed 7000S and I patched this part of the routine out inside the main exe file by altering a few bytes here and there. So now I can use the isopro SW with both machines.

Overall, it's very intuitive to use. However, you also need to know all the ways things can go wrong because there are so many ways to trash the PCB with unwanted extra scoring or mistakes with mirroring or alignment. You can also damage these machines quite easily and it's also very easy to break the tools with just a slight bit of operator error or clumsiness.

I prefer to use the LPKF circuitcam software to generate my isolation files and then I import them into Isopro. You 'can' do it all with isopro but the isolation/rubout generation algorithms are nowhere near as nice as the LPKF ones.

Also, the 1990s designed T-Tech 7000S machine is definitely not designed for operator safety. If you got your finger caught in the exposed  moving parts I think it will chew it off without even groaning or slowing down. It is very powerful. So don't use it if you wear a tie or jewellery and don't use it when you are tired.

So this is where the real experience comes in. You can only get this experience the hard way or from an experienced tutor.




« Last Edit: June 04, 2014, 11:01:20 pm by G0HZU »
 

Offline VK5RC

  • Supporter
  • ****
  • Posts: 2672
  • Country: au
Re: LPKF
« Reply #22 on: June 05, 2014, 03:29:19 am »
Thanks Mr S and G0HZU,  I suspect for my infrequent need it may not be worth the time to get it all 'right'  and $,  I appreciate your efforts on the forum explaining the process.It must be be very satisfying when it all comes out spot on :-)
Whoah! Watch where that landed we might need it later.
 

Offline s103_protomatTopic starter

  • Newbie
  • Posts: 6
Re: LPKF
« Reply #23 on: July 31, 2014, 04:11:27 pm »
Hey all I am working on documentation for all of our equipment listed in the initial post. I have mostly completed it for the CircuitPro software/S-103 milling machine and added them as attachments. I hope you find this useful but if you think I can explain things a bit more I have no problem with that. I am continually updating these documents so I may repost them in a few weeks. LPKF also has a ton of manuals on the website.

We are currently working on the issues with the pressing procedure (multiPress). The process documented by LPKF is flawed and leaves air bubbles trapped between layers. These air bubbles allow copper to be plated (using miniContac) into the internal planes causing shorts all over the place. We have diagnosed this problem through various 4-layer test boards we made with hundreds of different sized holes spread out across the 9x12 panel using various isolations. We found that the only way to guarantee no shorts is to have a 15mil isolation area around through holes  on the internal planes. We made 10 test boards using 8mil, 10mil and 15mil isolations and found zero shorts on 15mil in every test. only once did a board not have a single short. As a work around we have started messing around with vacuum bags which has yielded perfect results. We have made two test boards using the vacuum during the prepress stage and 5 minutes into the main press stage before our bag starts to leak. Both of these boards came out without a single short and we are currently designing a new test board to find the smallest isolation we can use reliably. In addition we are sourcing different vac bags that can take the heat and pressure from the press and may even be reusable. I have not documented this process yet but once we have a nice procedure I will post it on here as well.

I have included pictures and an excel file with our compiled results. The close up picture of the hole is the inner plane of a board I took apart for inspection and is magnified 10x. Before this we thought the milling machine was leaving copper behind or the holes were not centered in the isolation. Obviously neither is true and we finally realized it was an air bubble problem. I also included a picture of a test panel we made to give you a better idea of what we did. My suggestion moving forward is to design boards with 15mil isolation on internal planes unless tighter tolerances are needed in which case I can just setup a vac bag.

For those of you that are curious as to why we use this equipment. We typically make 50 boards at a time to save manufacturing cost. If it is a new board that has yet to be tested and all 50 come back with a design flaw your out 50 boards. We can turn around any two layer board within a day and any four layer within a week for $100-$300 which makes this equipment worth it especially now that we can reliably make boards up to 8 layers (6 is the most we have tested).

Still hate this equipment but when I can get everything working perfectly it is actually pretty sweet being able to make a multilayer circuit board in our tiny E-lab.

PS. We have recently had a board catch fire due to a pad coming off the board and shorting to ground (we think). The board had been in a prototype of ours for a few months and failed. I cant exactly blame LPKF but we will never use one of these boards in a customers product after this. Still great for rapid prototyping.
 

Offline s103_protomatTopic starter

  • Newbie
  • Posts: 6
Re: LPKF
« Reply #24 on: July 31, 2014, 04:13:07 pm »
Documentation
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf