WORKSHOP VIDEO #54: 983A/986A Solder paste dispenser

Without a doubt the solder paste dispenser is a cool tool. Compressed air hisses and solenoids crackle while precise amounts of solder paste extrude into beautiful droplets on a PCB. A few minutes with a paste dispenser saves $15-$25 (or more) on stencils, a life saver if you want to pick and place a one-off prototype. Only Akiba’s CNC milled stencils seem to beat it in speed and convenience.


Our dispenser is a random-name 986A, which seems identical to the more widely available 983A. There’s dozens of brands that all look similar, and generally use the same model number as a Hakko tool. Qwik, Best, Aoyue, Gayoue, Qinsi, Gold, the list goes on and on. Check it out in the video above, and continue reading about it below.

Last week we unboxed a load of tools shipped back from the parts market in Shenzhen. We’re assembling a mini SMD production line with a bunch of the tools we spotted in hackerspaces and workshops. While we don’t plan to produce anything for sale, we hope to learn more about manufacturing and possibly ease our prototype assembly backlog.

Over the next couple weeks we’ll demo each tool. In April it may be possible to get your own through a group buy, but we’re also scheming with Seeed Studio to make things directly available from their store.


First up is the solder paste dispenser. It’s a relatively simple tool that provides controlled bursts of air pressure to a solder paste syringe with an airtight piston.

The controlled bursts make sure that an exact amount of solder paste is released each time. It’s triggered either by holding a button mounted on the syringe, or via a foot pedal switch.968A-solder-paste-dispenser

There are plenty of settings available to find the sweet spot for your mix of flux and solder. For “Mechanic Paste“, a cheap tub available on eBay for a few bucks, we’ve been using 50 psi of pressure and a 0.08s to 0.12 second dispensing time. We also use a tiny bit of vacuum to help separate the paste from the syringe, but not too much because it seems to bleed air to make the vacuum.

On the back 4 DIP switches set the dispensing mode. Some modes give 1-16 drops per button press, at an interval specified by the interval setting on the front. A continuous mode dispenses long paste snakes – maybe a good setting for laying paste for ICs.

The instruction suggest holding the syringe at a 45 degree for optimal dispensing. Then hit the button or pedal and paste shoots out according to the mode set by the DIP switches.


We tested the paste and reflow process by tossing a bunch of different sized parts on one of our QFP v2 protoboards, available at Seeed for $10. We added 0805 capacitors, 0805 chip LEDs, a 64 pin TQFP PIC16F, a button, and a USB jack. A variety of parts with different footprints and thermal mass.

Everything went into the oven for our tweaked heating profile, more on that soon. Most parts aligned properly and soldered nicely to the board. The TQFP PIC had a bunch of bridges, certainly caused by lack of experience and easily fixed with some solder wick.


Three syringes, 6 plastic nozzles, and 6 metal needles came with the dispenser. We’ve had no luck getting any gauge of needle to work well, and have settled on the mid-sized blue plastic nozzle (in use, not shown).


70 to 100PSI of oil-free, dry air is needed to push the paste out of the syringe. While the dispenser may be cheap, enduring air compressor ownership could be hazardous to your health.

We found this 6L cheap compressor that’s slightly smaller than a tower computer for around $100. 24L seems much more common, but they’re way too big for our workspace. If you get a cheap one like we did, expect lots of noise. Several commenters have suggested a second hand low-noise dentist’s compressor as an alternative.

The air should be run through a dewatering stage. We’ve ordered one but it’s not yet arrived.


The dispenser air in and air out are both connector-less press in, pull out quick releases. The compressor-side connector though is something much larger than the standard 1/4inch quick release air hose we’re familiar with.

No adapter was available locally so a hack was employed. The end of a standard air hose was tenderly chopped, then the dispenser hose gracefully epoxied inside. A dash of hose clamps for safety, and a wad of tape and bubble padding to lessen the blinding factor of the hose clamps should this ever blow apart and whip around the shop.

You can check out previous workshop videos here.

Join the Conversation


  1. It doesn’t seem like it would use all that much volume of air, so I wonder if a second hand 80cf scuba tank & regulator wouldn’t be a possible way of offloading the noise onto a dive shop. I suppose it would be more expensive both upfront ($200 tank) and in continuing costs (annual inspections & refills), but you’d save on the air drying stage and ear plugs.

  2. Have you considered a CO2 canister for air pressure? Not too expensive, refillable for way cheap, would last forever in this application, perfectly constant pressure (no compressor to make it unsteady), and CO2 is hydrophobic, so guaranteed dry…

    I think my setup for CO2 (for other uses) was $150 including (smallish) tank and regulator, costs $20 to refill, and lasted me for literally years of airbrushing and other around-the-shop uses.

  3. Any chance we could get a peek on the inside? I’m designing a dispensing stage for my manual pick & place and would love to see how a more commercial one is put together internally.

  4. Use of CO2 indoors can be dangerous, I for one wouldn’t store a bottle of CO2 in my apartment….In use it wouldnt be an issue but a leak could prove hazardous. I wonder if a cheap 5 gallon compressed air tank you fill up at the gas station would run the solder paste dispenser for a usable period.

    Here are some CO2 exposure levels:

    The U.S. EPA CO2 exposure limits: The U.S. EPA recommends a maximum concentration of Carbon dioxide CO2 of 1000 ppm (0.1%) for continuous exposure.
    ASHRAE standard 62-1989 recommends an indoor air ventilation standard of 20 cfm per person of outdoor air or a CO2 level which is below 1000ppm.
    NIOSH CO2 exposure limits: NIOSH recommends a maximum concentration of carbon dioxide of 10,000 ppm or 1% (for the workplace, for a 10-hr work shift with a ceiling of 3.0% or 30,000 ppm for any 10-minute period). These are the highest threshold limit value (TLV) and permissible exposure limit (PEL) assigned to any material.
    OSHA CO2 exposure limits: OSHA recommends a lowest oxygen concentration of 19.5% in the work place for a full work-shift exposure. As we calculated above, for the indoor workplace oxygen level to reach 19.5% (down from its normal 20.9% oxygen level in outdoor air) by displacement of oxygen by CO2, that is, to reduce the oxygen level by about 6% (1.4 absolute percentage points divided by 20.9% starting point = 0.06), the CO2 or carbon dioxide level would have to increase to about 1.4% 14,000 ppm.

    1. If an industrial gas cylinder is leaking… That’s a problem! I doubt enough is released in this kind of use to cause any sort of problem – but you’re right, the potential is there… Hmmm… I wonder if anyone makes an affordable O2 concentration sensor?

      1. Its one of those things that can and does happen. For example you forget to close the tank valve at the end of use, meanwhile the solenoid is stuck open and fills your apartment with a dangerous level of CO2 over the course of a day or two. A woman was recently killed in a McDonalds bathroom when the air in the restroom reached a dangerous level of CO2 due to a leak in the bulk fill line for their beverage system.

    2. As an offtopic comment I would say that keep the CO2 levels below 850 ppm because this is recomendation for the office buildings where people who use brains to earn money for the owners are wanted to be the most productive. Inversely the common environment in submarines is 3500 ppm CO2 in the air.

      Another thing is that CO2 sensors are now relatively affordably available.

  5. A few of us attempted to use one of the solder paste machines in conjunction with a Mikini CNC mill to do automatic solder paste dispensing. What we learned is that it is easier to push solder paste through a thicker needle than a thinner one. While the CNC machine was accurate enough to position the needle over a .65mm pitch part, we could not get it to work with a needle small enough to leave the solder paste just on the fine pitch pad. The solder paste was too viscous to push through a fine needle. The Solder Paste Dispensing Using CNC. Just adding some information that people might find interesting.

    1. That parallels my experience trying to dispense by hand with needle tips. I did question my solder paste because it was a few years old, although refrigerated the whole time.

      I’m guessing now that the needle tips are best for dispensing glue or icing onto dollhouse sized decorated cakes.

    2. Looking at the video, you have ~1cm needle tip. Sure, it may be quite a problem to push viscous paste thru such length (and limit inertness). Cutting it to stick ~1*mm* gives much better results.

      1. That is a really good idea. I wonder why I never thought of that. I guess I thought vendors would ship the needles with the appropriate length, but there is no reason to assume that they would. We’ll have to give that a try.

  6. The only dispensing setup of this model that I’ve found on AliExpress is this (same as in the video) one for $128.

    In the end, this really doesn’t seem to offer that much more than a nice case with a gauge that’s just a timed solenoid valve and the custom dispensing handiwork. Given all of the work that one has to do to get it to mate with a pressure source, I’d be hard-pressed to fork out $128 for what’s only two steps up from a DIY dispenser. For half that, yeah probably. In between, eh, not sure.

    Gotta admit that having an automated dispenser would be nice to have. It’ll be interesting to see what/if the Seeed folks come up with. All the better if they can pair it with connectors that make sense for the pressure sources that we’re likely to come across in the West.

    Hell, if the price is right I’ll pump it up myself with a presta/schraeder-valve bicycle pump.

  7. Have you tried using extra flux on the TQFP pads? I’ve found that even though the paste has a fair amount of flux in it, putting some extra on seems to make it reflow more cleanly.

    1. To solve which problem, the bridging? That might just work because there’s enough extra tinned trace-length for the excess to run away. Otherwise for regular pad sizes, I’m not sure that it would, as there simply wouldn’t be enough extra space for the solder to run.

      The real question is if the extra flux will last long enough during the soldering cycle to matter. I think that that’s the part that I’d be concerned with not working as hoped.

  8. That connector will fit in a 3/8″ quick disconnect. You didn’t need to chop it.

    I can’t believe you have to use such high pressures and that you can’t get the needles to work still. I typically use the chipquick standard leaded paste, but have used the large tube sparkle paste brand too without issue.

    For a compressor, I have used both a large 4 gal at home and one of these at work (it’s pretty quiet!):

      1. Yes. The Sparkle Paste was properly handled and stored cold, I warmed it to room temperature. The Mechanic Paste had never been stored properly, either by the seller or myself in all likelihood.

    1. Thanks for the info. I only chopped the very end, I still have most of the original cable with the connector and I’ll look for an adapter now.

      1. I will say it is a snug fit, but it will work, and no air leaks out. I have a 1/4″ quick connect going to a 3/8″ female threaded quick connect (don’t forget the teflon for the threads!

  9. Whats the standard clip on airtool fitting on the clear hose? Is that one you got after you did the epoxy/bubblewrap/tape job? Or am I mistaken and the dispenser has a strange socket that the connector wouldnt fit into?

    1. The clear hose is for the dispenser. It came with, I now believe, a 7.5mm Asian style high-flow connector. The air compressor has a 1/4 inch female “industrial” style quick connect. I got a air hose with a 1/4 inch male quick connect, cut off the female end, and epoxied the tube of the dispenser into it.

      I would say this is the best match, and it measures 7.5mm inner diameter exactly:–Nipples/Series-315-DN-75/

      This odd asian coupling is probably why I couldn’t find any adapters or replacement fittings locally. Could be an issue for anyone who wants one of these, we’ll have to work that out.

  10. Have you tried to see what the min usable air pressure is for the different pastes you have. If so would it be possible to use an air brush compressor. they run at about 30-45 psi max and are almost noiseless.

  11. Well, Ian, you sold me. Just bought two of them from AliExpress, along with two tubs of mechanics paste. We’ll have to see if it works as well for me…

  12. Does anyone know what hardware they use for creating the suck back feature of a soder paste dispenser ? I have done some research and realize a pneumatic system is never going to be as accurate as a ballscrew driven unit but the simplicity of a pneumatic system is attractive. I am interested in building my own and have considered (but not tested) a vortex (vacuum generator using air pressure) for suck back. I am concerned that air volume variation in the syringe from full to empty is going to be an issue. Any thoughts ?

  13. Hi!
    Great video! It inspired me to buy one to test out.
    However my chinese is not really that great and my manual is in chinese only. Anyone knows about the different dip switch settings?…or even better…have an english version of the manual?

    Best regards,
    Fredrik, Sweden

  14. I’m in the same spot as Fredrik… My units arrived this week, but the manual is greek, er, mandarin to me…

  15. Good & informative video was interested to see someone have better luck with pneumatic dispenser. We have been using hand applied since our 0402/0603 parts don’t work as well with dispenser, the paste drops must be very small. Looks like you have the same problem with PQFP/TQFP parts that we had with the even more annoying QFN parts which have no exposed leads so hand touch-up is difficult at best. I have good luck dropping paste like you did but then I chase the paste with an iron to tin the footprint apply flux and IC then bake, it takes longer to fix afterward then to control the solder deposition up front & these shorts are the result of excessive solder on footprint extra flux helps but not much.

  16. Check out this link:

    It is a complete unit that can be plugged into a 12 volt supply and is ready to go. Uses NO AIR. Takes standard off the shelf syringes that come pre-filled with solder paste. Can control drop size and flow with the unit itself. Very ergonomic, and self contained unit.

Leave a comment

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.