I recently bought my first hot air station. On a whim, I decided to get the Xtronic 9020. It comes with a soldering station and fume extractor.

After a week with the station, I figured out that I have no use for anything but the hot air. I decided to change the soldering iron out for a desoldering gun.

Step 1: Items Needed

First, I did some testing of the vacuum pump to see if it would work. Unfortunately, my testing indicates that the diaphragm pump in the station does not move enough air to power a desoldering iron. It could not clear a plated thru hole. This hack would have been so much better if it could have used the existing air pump!

5 minutes of googling, and I was watching Dave teardown a Rhino desoldering station on the EEVblog. Inside, there was a large pump that looked very similar to images I received when googling "12V vacuum pump," only bigger. I ordered a 12V vacuum pump and a desoldering gun, and I would see what I would see. As it turns out, this pump is a lot better than the internal diaphragm pump, but it is still lacking in power - at least when run at 12V. It will clear solder bridges and empty plated thru holes, but it is not always strong enough to cleanly desolder thru hole parts. See last page for additional info.

1. Desoldering gun part 88-552A, replacement desoldering gun for station ZD-985 station. This cost me around $45.00. http://www.amazon.com/88-552A-Desoldering-gun-ZD-...

2. 12VDC vacuum pump. I have seen these on Ebay for as little as 12.00. I purchased mine on Amazon for the Prime 2 day shipping. http://www.amazon.com/88-552A-Desoldering-gun-ZD-...

3. 1x 4.25K smd resistor, 0805 or 1206 size will do.

4. 2x 10K resistor, 0805 or 1206 will do.

5. 1x 2A N channel Mosfet

6. 1x 3904 NPN signal transistor

7. 1x 1.5K resistor

8. 1x flyback diode for the motor. I used a little 914 signal diode.

9. 1x 270uF 25V electrolytic capacitor, or larger.

I had a soldering iron holder from an older station I no longer use. And I also needed to use a bit of brass tubing to make a 90 degree elbow for the air hose. That's about it for parts. Aside from the first two things on the list, I had all the other parts lying around, already.

As far as tools, I used a hand drill, pliers, phillips screwdrivers, soldering iron, and a hot glue gun.

Step 2: The Connector Pinouts

Ok, the first thing I did was to determine how I was going to plug the thing in. Luckily, the soldering iron on the 9020 has a 6 pin plug on it, because it has an internal magnetic field sensor in it. The plug on the 88-552A soldering gun is almost identical, screw on DIN connector, but with 7 connections. I searched digikey and mouser for a replacement plug, but I had no luck. No worries. I would just take the plug off the soldering iron. I figured I could find a way to make the soldering gun work with just 6 connections.

If you unscrew the little phillips head screws on the connector, you can pull it back and expose the solder terminals on the plug. This, I did. Then with a scope and a DMM, I was able to suss out the following: Soldering iron 6 pin connector:

1. Red wire - heating element

2. Yellow wire - heating element

3. Black wire - magnetic field sensor open drain output; on station end this connects to a digital input pulled up to 5V

4. Green wire - Vss: magnetic field sensor ground, thermocouple negative

5. Blue wire - thermocouple positive

6/center. White wire - connected to metal tip of the iron for earth grounding

The desoldering gun was a little trickier, because I didn't have the station to power it on. So I had to guess a bit more. Other than initially mixing up the thermocouple + and -, I got everything else right the first try! Desoldering gun 7 pin connector

1. Black wire - thermocouple positive

2. Blue wire - Vss/thermocouple negative (see why I mixed those two up?)

3. Red wire - heater element

4. Red wire - heating element

5. White wire - trigger switch contact, NO press-to-make switch

6. White wire - trigger switch contact, NO press-to-make switch

7/center. Mysterious part. I removed the cover and googled the part number. It turns out it's a vibration sensor. No doubt it's used for an automatic sleep function.

The desoldering gun also had an earthing wire connected to the shield/connector itself that is in continuity with the tip. If you are counting, that's two more wires in the desoldering gun vs the soldering iron. After ditching the vibration sensor output, that's still one connection too many. I decided to solder one end of the trigger switch to Vss, and the other end to connect to pin 3, so that the trigger would double as the sleep sensor for the iron. And now I simply desoldered the plugs and resoldered the 6 pin plug onto the desoldering gun's cord. The screw on termination/strain relief parts were practically the same. I did not even swap those, although I needed to file on the 6 pin plastic part a little to get it to fit.

Step 3: Futzing

The thermocouple on the soldering iron measured 50 ohms. The thermocouple on the desoldering gun measured only 4 ohms. So I figured there would be a different voltage output, and I wasn't disappointed. The desoldering gun measured only a couple degrees C when I turned on the station. And by the time the temp reading crept up to a few degrees, I ended up accidentally burning a hole in the top of my work bench, playing with a bead of solder. lol.

So at this point, I had to remove the pcb from the station to see what jived. I quickly located the lm358 opamp to which the thermocouple output connected, but I have to admit I can't figure out the circuit. I'm more into the digital end of things, and I have to admit I don't know my way around opamps too good. I ended up using trial and error to get a combination of resistors that would make the thing work. See pics.

The end result is that the iron measures 104C at room temp. This is good, because 100C is the lowest that you can adjust the station. So I can hit the bottom end of the range for sure. Also, the automatic sleep function (more on that later) kicks the iron down to 100C. So it will completely shut off the desoldering iron. Bad news, is the iron still gets working temp at only 120C. So there aren't as many steps as one might like, here. But that's fine with me. 15-20 steps is plenty enough for desoldering work, in my estimation. I suppose if I switched it to Farenheit, I would have 9/5's the resolution. :)

Sleep function: I sussed out that the sensor in the original iron has an open collector output. The input to the station is pulled up to 5V logic high. When the iron is placed near a large iron object, the sensor pulls low. When the sensor is low for the duration of the sleep timer, the iron goes to sleep. But I also noticed that the iron will go to sleep if the sensor remains HIGH for the duration of the sleep timer. In other words, the sensor must trip occasionally to keep the iron awake. I figured attaching the trigger to this signal makes perfect sense. If the trigger isn't depressed for the interim, the gun will go to sleep.

Step 4: Putting It Together

Well, I hoped that I could tap power for the motor from the station's psu. I wasn't totally disappointed.

****update 072214 : I have added an external, compact switchmode 24VDC power supply to the station to power the 12VDC vacuum pump. This boost in power makes the iron work very well. I leave the rest of the instructions unchanged for posterity. (If you are going to use a voltage higher than 20VDC to power the motor, doublecheck the maximum Gate-Source voltage rating of your N Channel MOSFET; a lot of them top at 20V. You will need to add some circuitry or a new mosfet that can handle the voltage.)********

There's is a 24VAC output on the transformer for powering the iron. There's also a 10VAC output for powering the control board. This is rectified on the station's pcb, by a full bridge rectifier made of discreet SMD diodes. The result is a 15VDC float voltage with enough juice to power the motor at a full 12VDC (when the heater element isn't on, at least!).

Small problem, though. Starting the motor sometimes caused the station to reset itself. So I looked in my capacitor box, and I found I had a few 470uF 16V caps. I put two in series to make a 235uF 32V cap, and I added these to the existing 1000uF filter cap already on the board. Hallelujah. This 20% increase was just enough to do the trick.

To be sure, the motor does not run full strength while the desoldering gun is heating up. The transformer just can't power both at the same time, even though they are powered from different taps. But once the desoldering gun is up to temp, it works fine. Once to temp, the power to the heater goes on/off only very, very briefly, and this does not seems to adversely affect the vacuum function.

I initially tried to just stuff the motor into the right side of the station, where it just barely had enough room. But one of the heatsinks (for the soldering iron triac) was in the way of the air hose, causing a nasty kink. I found the best recourse was to move the transformer a half inch to the left, necessitating drilling two new holes. And even still, I needed to make a 90 degree elbow to get the hose to connect without kinking. I was careful to leave enough room so that when the diaphram pump is on, it would not hit the motor, then I hot snotted the motor into place. So now, the vacuum pump pulls through the connector on the front, and its output is unattached. It just blows into the interior of the enclosure. Likewise, the input of the diapragm pump that powers the hot air is left unattached. It just pulls air from the interior of the enclosure.

Now I just needed to rig up some transistors to get the switch to power the motor. See the pics for a schematic. I built this up on a little bit of protoboard and hot snotted it onto the controller pcb.

*Note, in the schematic, I show an 8k resistor on the base of the npn transistor. In the parts list, I have that as 10k. It's no matter. It would probably work even without the resistor, since that end of the trigger switch is wired to the sleep signal on the station. The pullup on that input pin probably gives enough juice to drive the transistor, by itself.

ExpressSCH.pdf
Download

Step 5: More Info

Disclaimer: This hack works on MY station. Due to potential firmware and hardware revisions, it is not reasonable to expect that this will necessarily work exactly as indicated on your 9020. OTOH, if you know what you are doing, it is perfectly reasonable to expect you can get nearly any brand desoldering gun to work with nearly any soldering station, as long as the psu is strong enough to do the job.

***UPDATE***07/20/2014

I have never used a desolder gun before, and I found out my technique was wrong. I wasn't letting the iron get hot enough, and I was clogging the gun. The pump might be borderline strong enough as it is, once you know what you're doing. But to get better performance, I have added an external psu and am powering the vacuum pump at 24VDC. It works. It works, well. If the pump motor burns out prematurely, I will update.