Pick and Place for small batches

[michaelym]

Hi all,

I saw the Neoden 4 and was wondering if there are better machines that can be recommended in that size and price range

Most of my parts are in the size range of 0603 and my PCB's have 100+ unique parts (I understand that I would need to run them through more than once)
I do batches of 20-40 units at a time

Thanks for your help

[mrpackethead]

if thats your situation, you probalby would want to

(i) look at ayour design really carefully and see if you can reduce the BOM.   
(ii) Get two machines at least.  You dont' want to be running half a board, stripping down and setting up for a second go.

[mikeselectricstuff]

Hi all,

I saw the Neoden 4 and was wondering if there are better machines that can be recommended in that size and price range

Most of my parts are in the size range of 0603 and my PCB's have 100+ unique parts (I understand that I would need to run them through more than once)
I do batches of 20-40 units at a time

Thanks for your help

It wouldn't be practical to do multi-pass on a single Neoden4 as swapping feeders looks very fiddly.

(i) look at ayour design really carefully and see if you can reduce the BOM.   

Absolutely. Look at how many resistor values you could eliminate using series/parallel combinations, how many different cap values you really need etc. Any additional parts cost would be more than made up by savings in feeders.

[SMTech]

^^ what he said. Having said that after you've loaded and setup a few boards like that you should soon realise on small batches how much of the cost if you were subbing out was going on setting up.
Stripping down and starting over would require careful planning, you would need to run through your parts in height order and kit them to yourself split into the mini jobs, way to save a rummage fest for each loadout. Every extra line on a BOM is another component to buy, another component for RS/Farnell/Mouser to pick incorrectly,another part to load, another part to unload, another part to store and label all of those things take time and if you're working in small qtys and strips of tape come with a massive added risk of getting mixed up. Parts that come in larger tape sizes also use up proportionally(usually) more "slots" on your placement machine, 48 feeder slots really is not very many. As Mike points out the neoden feaders look fiddly (essentially fit once never move, I wouldn't want to be moving those terminals that often), so one of the rival machines that uses a more generic removable style might be better.

Consolidating the BOM: the first thing I would look at is pointless "cost savings", in low volumes the component cost of passives is basically irrelevant so if you have a bunch of special values look at why.
Examples of this can (IME) include having the same value cap in multiple voltages or the same resistance but some are 0805 to handle a power requirement. Next you can look at weird resistor/cap values - maybe you can make some of them from more standard ones already in the design. I would go back to the basic design itself last, check the specs/tolerances, arbitrary pullup/bypass choices etc.
More extreme measures could include splitting out bits of the board as modules, handy if you frequently reuse bits across multiple designs and can sometimes make it easier to test/fault find.

[mrpackethead]

I was thinkingt that once you'd placed 40 boards on a Neo-den that the paste woudl probably be past its hold time, in order to do the next run.   
 
I run 150 slots + a big tray.  I'd hate to think about having to move to jsut 50.   

[SMTech]

Depends on your paste, mine has a board life of 3 days...

[mikeselectricstuff]

arbitrary pullup/bypass choices etc.

For example if you need an 11K to set a voltage on a regulator, all your pullups can also be 11K.
if you're using example designs from datasheets, look at every value and ask if it's critical - e.g. boot & softstart caps on buck regulators can often be replaced with higher values.

[mrpackethead]

Depends on your paste, mine has a board life of 3 days...

What paste would that be?...

[Kjelt]

arbitrary pullup/bypass choices etc.

For example if you need an 11K to set a voltage on a regulator, all your pullups can also be 11K.
if you're using example designs from datasheets, look at every value and ask if it's critical - e.g. boot & softstart caps on buck regulators can often be replaced with higher values.

Unless you need the 11k to be 1% or better you would put a 10k and 1k in series and use the 10k for pullup.

[mikeselectricstuff]

arbitrary pullup/bypass choices etc.

For example if you need an 11K to set a voltage on a regulator, all your pullups can also be 11K.
if you're using example designs from datasheets, look at every value and ask if it's critical - e.g. boot & softstart caps on buck regulators can often be replaced with higher values.

Unless you need the 11k to be 1% or better you would put a 10k and 1k in series and use the 10k for pullup.

No - that would be two BOM lines

[SMTech]

What paste would that be?...

LFS-UFP-T4 - a similar paste to GC10, however extended stencil and tack times are not that unusual, although if you push the limits, while it may reflow perfectly fine it won't be as "sticky" at the end as when it was nice and fresh. When we ran a much slower machine we would paste a whole rack of PCBs (manually) to get that process out of the way and that sometimes meant they remained pasted overnight, it was never an issue.

Worth pointing out anyone who runs a business that clearly has a requirement for solder should be able to get a free sample out of most manufacturers/suppliers if they request it, typically a 250g pot. One assumes they filter the requests a little but certainly helps with the cost of running trials.

[mrpackethead]

What paste would that be?...

LFS-UFP-T4 - a similar paste to GC10, however extended stencil and tack times are not that unusual, although if you push the limits, while it may reflow perfectly fine it won't be as "sticky" at the end as when it was nice and fresh. When we ran a much slower machine we would paste a whole rack of PCBs (manually) to get that process out of the way and that sometimes meant they remained pasted overnight, it was never an issue.

Worth pointing out anyone who runs a business that clearly has a requirement for solder should be able to get a free sample out of most manufacturers/suppliers if they request it, typically a 250g pot. One assumes they filter the requests a little but certainly helps with the cost of running trials.

Live and learn every day..  120 hour tack time.         GC10 is quite a lot shorter.  I tend to be stenciling just before it gets put in the pnp.   


I'd now need a compelling reason to move away from GC10, With our vapour phase, and stenciling set up, it just works and delviers us a result that is so much better than anythign we've done previously..   YMWV,   And you have to dial in your line and get all your stuff correct.     Its not somthign you can get out of a text book, experience will be your freind. so the only way to do it, is to do it.

[Geoff_S]

I'd now need a compelling reason to move away from GC10, With our vapour phase, and stenciling set up, it just works and delviers us a result that is so much better than anythign we've done previously..

I'm looking forward to a thread from you about your vapour phase oven

[mrpackethead]

I'd now need a compelling reason to move away from GC10, With our vapour phase, and stenciling set up, it just works and delviers us a result that is so much better than anythign we've done previously..

I'm looking forward to a thread from you about your vapour phase oven

What do you want to know.  Its a really simple process.

[Geoff_S]

Sorry, I didn't want to de-rail this thread.  I'd seen you mention in some other thread that you'd made a vapour phase oven and was just curious to see the physical setup - what you've using for the enclosure, the fluid heater, control system etc.

[Gary.M]

Go on Packet head... A diy thread with pictures. I bet it will be popular.

[mrpackethead]

Go on Packet head... A diy thread with pictures. I bet it will be popular.

Yeah, i bet it will be too.    I have partially considered commericaliing the idea. but its not our core bussiness and it might become a distraction.. I need to think about it a bit more.

however in essence if you are prepared for it to be 'slow' in comparison to a production line reflow, and energy inefficent then you can make vapour phase VERY VERY easily.

[Gary.M]

Ok, we're all waiting eagerly.

[Geoff_S]

Yep sounds interesting ! Slow and inefficient isn’t a problem for prototyping...

[mrpackethead]

basically its just a stainless steel tank.   Mine is 400mm deep, 400 wide, 600 long.   I had it made in stanless steel.  On the bottom side of tank  i have a ceramic heater element ( 1800W ).      Inside the tank we put about 4mm of Galden  LS230 in the bottom.   I have a stainless lid that fits on teh top and makes a resonably tight fit.

Heating is easy. You need a way to cool it as well, and to do that i pump cold water over the outside sides of the tank.   ANother system uses air fans, but that takes far too long.

boards go in the tank. on a little pallet.   Put lid on.   Heat cycle.   Galden heats and then creates vapour cloud.    THe main vapour cloud only rises about 50mm above the PCB, but there is a seoncary cloud circulates through the rest of the tank.  You dont' want this escaping or you'll use lots of ( expensive ) galden.  It condenses on the lid and falls back into the tank.       ONce the vapour cloud reaches 230C,  you stop heating and then start the coolign.

I know this does not really suffice, but its the guts of what im doing..    Pretty busy right now, but will give you a bunch more detail later..    You probalby can build the entire thing for a couple hundred bucks, ( or less if you can weld stainless ) .  The galden will cost you lots more than that!

[Kjelt]

It would be interesting to see the reflow profile curve.
The cooling stage sounds challenging in a closed tank

[mrpackethead]

The profile of VP is quite differnet to that of a a convection reflow;     This is a really good paper on some of it.

I do have a profile somehwere, i couldnt find my graph though.   At the moment i largely just heat the galden from cold to hot, without trying to do anything too clever.       It largely is a linear graph.

https://www.smta.org/chapters/files/Arizona-Sonora_IBL_SMTA_AZ_Expo_2012Dec4.pdf




Vapour Phase Soldering (VPS) is an alternative method for widespread reflow soldering technologies (e.g. infrared or convection type). VPS has emerged in the last few years after the introduction of Galden fluid. The vapour of Galden is used as the heat transfer medium. The main advantages of VPS are the lower peak temperature, nitrogen/oxygen-free inert atmosphere, the improvement in solder wetting and the reduction of the profiling time. One disadvantage of VPS is the non-controllable, linear soldering profile achieved by common VPS soldering method. With controlled dipping of solder goods into the vapour space, non-linear soldering profiles can be obtained. This method is called Soft-VPS (S-VPS) and was developed and patented by IBL. With this method, custom temperature gradients and profiles can be realized, reducing voids, tombstoning and avoiding high thermal stress on the boards. This paper presents thermal profiling of an experimental VPS station from the aspect of this method, with a successful optimization of the process. The optimized profiles for low melting point (138 °C) lead-free Sn-Bi solder paste were inspected. Soldering results of 0603 chip size resistors were then evaluated with resistance measurements and x-ray inspection.

[1276-2449-1-ND]

I do short runs too. Because my PNP doesn't hold many components and changing reels is too much of a pain to be feasible, I went through and redesigned all of our PCBs to reduce the number of components we use on everything. Now about 90% of components can be placed on our PCBs at one time (worst case -- more than half are 100%) without having to touch the reels. The remaining components get tweezered on right after placing.

[mrpackethead]

I do short runs too. Because my PNP doesn't hold many components and changing reels is too much of a pain to be feasible, I went through and redesigned all of our PCBs to reduce the number of components we use on everything. Now about 90% of components can be placed on our PCBs at one time (worst case -- more than half are 100%) without having to touch the reels. The remaining components get tweezered on right after placing.

For small runs, hand placing a few parts after everything else is done is quite a valid thing to do.  I often do this, sometimes if i have a part that is particularly difficult to place. or is large and chunky.