Preparing solder paste for syringe-style dispenser
I'll throw my tiny bit of experience in here, since I too bought that dispenser and a cheap-ass Chinese reflow oven to play with a while ago.
Does IPA work long term enough? Yes!:
First off: Yes, adding IPA to the paste works great, and if you close the syringe with a tip-cap (cheap plastic Leur-Lock caps aplenty on eBay and AliX) it'll stay in quite long enough to use up the paste.
Though I admit that I use 10cc syringes myself, with a home-made glass-fibre holder and more ergonomic switch, because the 30cc is too bulky for my preference. Again sets with rubber-ring-sealed mounts and syringes a-plenty on many sites.
And I replaced the foot switch as well.
EDIT: Why I use IPA at all
It is true, from the comments, that 5 to 6 bar, or 0.5 to 0.6 MPa, will push out very thick paste easily.
But when working with thinner paste tweaking the droplet size using the pressure is a lot easier. And also, the droplets, in my case, adhere much better to the PCBs when just a tiny bit of IPA makes it a tiny bit more liquid.
The Paste I use:
That said, I also use, in true cheap-bastard fashion (proto's need to be as cheap as possible, yes?) I also bought Chinese solder paste, of the brand "Mechanic". Little orange pots.
How I use that paste:
I then add only a few ml of IPA to the solder paste, inside the little pot, stir well and then pour it into the 30cc syringe with tip-cap, put an end-cap on the syringe (Not the floater that they come with to push down liquids, but a cap, also findable everywhere) and let it sit for a a bit to get most of it down to the tip. Some slush on the sides is okay, since that'll wash away next time and I don't use that syringe. It's basically just a tiny funnel now.
Once enough has accumulated at the tip, you use that syringe to fill one or more smaller syringes using the widest needle you have and the dispenser set to continuous, very low pressure air, with no floater. With the floater you can have higher air pressure, but it's a horrible mess to get it out again, because paste will splutter everywhere. I have tried it.
Put the floater on the smaller syringe, push it in as far as you can. If possible let some air escape past its sides. Set it upright a while. If the paste doesn't flow down very willingly in a well closed syringe, heat it up to about 40 degrees C (in the closed syringe the IPA won't evaporate). Once the paste is at the bottom, take off the cap and using a blunt implement, carefully push the floater down.
How much IPA?:
Now, the amount of IPA to use is very dependant on your personal style and preference, as well as needle size and type. I prefer a thicker paste to make nice round droplets so I really put in only a couple of ml of IPA and use the conical plastic needles. You can tweak thickness a little for various hole sizes. The conical tips come in very tiny and very large openings, so it's hard not to find something suitable.
If you have very liquid paste, it will run out nicer across the entire pad (if you have a thick solder mask) or not so nicely across your entire PCB (if you have no or very thin solder mask).
Paste type and thickness versus tip type:
If you dilute your paste very far, the conical tip is a bad idea, as it allows lower viscosity liquids to goop out uncontrolled, due to air-bubble exchange at the tip. Keep that in mind. With highly liquid paste you need a metal tip, but since it's not so viscous any more it won't add so much resistance either.
Will IPA change anything in the behaviour in the oven?
Possibly, yes. But. In the oven I use, with the normal 63/27 profile over 7 minutes I have found no difference in effectiveness or quality between with and without IPA after quite extensive testing on ENIG and HASL boards.
Do manufacturers make different viscosity paste? Yes.
Have I found one mfg that makes different viscosities aimed at different purposes to date? No. But I also haven't tried purchasing directly from a manufacturer who aims at industrial use yet, so chances are fair the practise does exist.
With a few hours of practise on mixing, filling a couple syringes and trying different tips after a while I found that I had a couple of nice notes about recipes for paste and the tips and dispenser settings that go with them that fit my personal preference down to 0402 / 0.5mm pitch chip mounting.
As a fun point of interest, with slightly more liquid, but not too much and a medium sized metal needle I found you can just lay a thin stripe of paste across a whole row of fine-pitch VQFN pads, on the outer edge and let it bleed in a tiny bit to then have the chip reflow nicely, since solder-mask repels the liquid solder, which prevents shorts.
The striping with relatively thick paste will cause unwanted solder balls, as will any kind on larger pin-gaps. I mainly use the trick on 1.5mm pitch and below. If then a few balls occur, use some IPA and a toothbrush to brush them neatly away.
Other than all of the above it will still take a few days of practise to get everything right for your own personal working style and I fully expect it to end up completely differently from any procedure I have described quite soon.
Re-Use of Syringes:
I opted to buy 100 sets of 10cc syringe + floater + cap and just throw the empty ones in a "recycling" box with the intent of cleaning and re-using them later. Possibly I will wash out the paste with IPA and let it dry to use again (if that ends up being possible at all). But I have only got 4 empty ones at the moment, so I'm not there yet.
The extra syringes also allow you to have 3 or 4 mixes handy for different kinds of jobs, also convenient, though I only use 2: A bit thick and slightly less, but still quite, thick.
Alternative to IPA
All said and done, there's one more option. Use liquid flux to thin the paste. They use a thick goopy paste, like the old rosin in those cans, but without the rosin content, to make the solder paste. In most cases this mixes quite well with standard none-water-based liquid flux. But be aware, most fluxes are either water based or alcohol based, which means it's just adding IPA in disguise, with a few additives.
I have not tried this for dispensing, just as a mixing experiment, because, hey, I have a 5 liter jerry can from Multicore anyway.
Air pressure and air cleanliness:
The compressor you're using can be set to nearly any pressure the dispenser can handle (up to 8bar at least, as I have tested this for a good duration), as the front regulator of this unit, in most cases works pretty well down to 0.5bar (I believe that to be <10PSI, though I'm not sure, because I find that a silly unit).
It is however somewhat advisable, for continuous correct operation of the regulator in the unit and the switch over long time, to have a moisture and small particle filter before the dispenser.
Because I work on Waver-Stepper parts as well for some customers, my set-up includes a 100micron filter and water-catch, followed by a 5micron filter, with one track going straight to my bench after that and one track going through a venturi-based -10degrees C water separator. (With the option of adding a low-flow-rate 0.5 micron filter locally for the most sensitive of devices, but I hardly ever use that for anything)
This makes me quite confident the thing will keep working for a very, very long time as long as the parts turn out to be at least moderately okay (only a few months in).
But, I'm also quite happy to say that I suspect just the 100 micron filter and water trap (that stage is usually combined in a single unit) would probably suffice.
I suspect any of these will help you keep slightly cleaner air, if you desire, at Chinese price and quality, of course:
Water trap with pressure regulator and gauge, likely around 100 micron
40 micron Filter
Inline Chemical Dessicator - Will take out more water, after the first water trap, but uses a sort of silica gel and will need regular refilling or drying
-- I did not search very well, they may be available cheaper.
Edit More:
I am now in the process of changing the unit from it's "built in vacuum" to an actual vacuum pump hook-up. The improvised venturi they use inside wastes a shit-load of air to create only a tiny vacuum. Since I have a vacuum pump, I only needed a vacuum pressure regulator extra, which I have just ordered. So soon my compressor will remain a lot cooler under the operation of this device.
The vacuum regulator is needed, because the main 0.6bar-below-room-pressure I normally have the line regulated to is strong enough to suck the plunger back out of the syringe :-D.
While some have recommended the use a stencil over dispending paste, there are certain trade-offs where your system will be faster (and cheaper... although stencils are very cheap depending on the complexity) than ordering and applying a stencil. For example:
- no extremely fine pitch components
- a stencil usually requires cleaning and set up times that can be avoided by simply dispensing paste onto a handful of boards
Before I answer your questions... you said:
adding a little isopropyl alcohol helps soften it enough to work "good enough" with this system
I use isopropyl alcohol to clean flux from my PCBs. I've never heard of anyone mixing it with their solder paste. I wonder if there will be any negative consequences from adding alcohol to your paste mixture. Is the reason you are adding alcohol to make the paste "thinner" so it comes out of the machine? Have you tried simply waiting for the paste to warm up?
That being said, from the way you have worded your question, it appears you very detail oriented. This is good when it comes to soldering and handling paste as it can be a scientific process and almost an art. You will find that it is also very forgiving. As you go along and gain more experience, you won't worry about things like this (at least for doing things by hand and making prototypes... mass production is a different game).
To answer your questions:
- Do manufacturers make different types of paste for syringe application vs stencil-and-squeegee application? The paste I bought from the local shop is by MG Chemicals, Sn63/Pb37, and comes in a syringe, so I assume it's the right kind for my needs.
No. They only make different mixtures of Sn/Pb/other materials. You'll find that your paste should work for most applications.
- What's a good ratio of alcohol to paste (assuming Sn63/Pb37)?
You'll have to experiment with this on your own. Even if someone were to recommend a ratio to you, the alcohol may dry up over time. Also, the consistency of the solder paste will change over time as well as it gets more exposure to air.
- Once I mix the paste and alcohol in a separate container, what's a good technique to get the paste into one of the syringes without smearing it all over the inside wall? I've thought about pouring all of the paste into the dispensing syringe and adding alcohol on top of that, but I'm afraid stirring within the syringe won't mix the mixture homogeneously.
Again, I've never heard of anyone mixing alcohol into their paste. If you were to do this, I would guess it would need to be in a larger container and then transferred to the syringe so it was indeed homogenous.
- What's a good tip to use on the syringe? The dispenser I bought came with a large sample pack of various types (metal and plastic), but none worked great. The paste is too thick to fit through the metal ones and I find the opening of the plastic ones to be a little too large to be precise.
This will depend on the size of your components. If you are doing anything more complex than a QFP package, you'll most likely need a stencil. You might even need a stencil for a QFP (or just be willing to do some re-work with solder wick by hand). You'll find that any reasonably small tip should do the job.
Overall, my advice is to get started with what you have. Your own experience and familiarity with your system will answer a lot of questions for you.
Here are some videos that have dramatically increased my efficiency:
- Random tips from Mike's Electronic Stuff: https://www.youtube.com/watch?v=pdGSFc7VjBE
- Felix Rusu's DIY pick and place $20: https://www.youtube.com/watch?v=YhavXauuWqY