ZD-939L Hot Air Station - 2 minute teardown!

[Wirehead]

I recently purchased a new hot air station, and seeing there's not a lot of info or user experiences floating around, I decided to take this apart.
I'm no expert on hot air stations, nor will I pretend to be one , but I kinda like this station so far. It gets parts of multilayer boards off quite easily. Oh yeah, it also seems to be doing it in a safe way (grounded exposed parts.. no Atten-drama here )

It was bought from a EU-vendor (Reichelt), semi-rebadged from Zhongdi (http://www.china-zhongdi.com).
I guess due to this, and German laws, they let them build this with a bit more QC.


The unit itself looks nothing special. It's a hot air station, what do you want? It has 2 buttons on the front. The right one is the main power switch, left one turns the heater on. After soldering, if you turn the heater off, but leaving the main power on, it triggers the pump to it's maximum speed, for about one minute, until the unit is fully cooled down. I think that's a nice feature. (Oh, don't worry about the funny colors on the LCD, that's due to a polarising filter on my lens...)


The pump is secured during shipping with transport screws. Once removed, the pump can move freely on silicone rubber mounts.


On the back of the unit, the power inlet has clearly marked ratings, as well as the name of the EU-vendor.


Connections at the power inlet are made with spade connectors, and are insulated. Grounding is provided with a shake proof washer, bolted to the housing. Another wire extends from this bolt, to the heater tip.


At the front, connection to the switches is made in the same fashion - insulated spade connectors. Note that for the main power, the Live (or Hot) wire is switched. For the heater, both wires are switched.


Note that all connectors have some glue on them, to make sure they don't get disconnected during transport.


The LCD driver is made by Holtek, the main MCU is a On-Bright OB39A16M1. Google Translate Datasheet. You can also see a J-EXT header (I suppose JTAG)


There's also an ICSP header. This and the previous option should leave open the option of custom firmware.


Overall, it's a nice little unit, that seems well built. (Notice thread locking on the bolts. The cover itself also has shake proof washers). It's not going to compete with a Hakko or anything, but for the hobbyist, it's safe, and it works.

[Arhammon]

This stations doesn't have thermal switch or fuse, typically.

[yngndrw]

There is a fuse, you can see it in the third picture - Built into the IEC connector.

No idea about the thermal switch, the heater's in the handle.

I have the ZD-939A which I purchased from CPC many years ago and is very similar to this but with analog temperature controls, works well.

[Arhammon]

Yes, I mean thermal fuse.
I also have 852 station. I suppose all Chinese compressor type stations similar.
It work well for the price.
But I saw how flame up 702 stations with fan in a handle. Two times.
Not funny video about this stations:
https://youtu.be/m5kzyUFb2Sg
https://youtu.be/XwJdYxe_HFQ
Thermal protection it's important feature.

[Wirehead]

Well, that's different. The squirrel fan stations are more prone to that kind of failure, because all the electronics are in the handle, and are exposed to heat.

Here, all the electronics are in a cool housing. I'll check what happens if I disconnect/short the temperature sensor. Normally, a thermistor fails open. So if that's the case, I'll see what the electronics will do.

Also, the membrane pump is dual membrane. Failure of one membrane, will not cause all air to be lost. The temperature will rise, and the heater will be turned down..

edit: also: who leaves hot air on, when you leave your workplace? Not a very smart thing to do, same with a soldering iron.

[Arhammon]

It's same - electronics in the main block, in the handle only heater and fan.
In our case it was triac failure. Lukey 702 station on video doesn't have mechanical power off.

[Wirehead]

Allright, so I checked the protection system. If the thermistor goes open (due to dropping the handle, old age, or e.g. excessive heat from triac failure), the display states "error" and does NOT heat the element in the handle. It's really a dedicated thermistor by the way. There's 5 wires going to the handpiece: 2 for the heater, 2 for the thermistor, 1 for ground.



The pump stays on in this case by the way.

[frogblender]

Has anyone measured the actual output temperatures on these?

I just finished measuring my ZD-939A.... what a total piece of crap - complete GARBAGE.

My ZD-939A is a few years old, it never worked very well.   Now I know why:  it has HORRIBLE temperature control - the output temperature oscillates by OVER 100°C.  wtf???    Here's my data from stuffing the thermocouple into the nozzle (trying not to impede airflow):

Set temperature:    365°C   Actual:    oscillates between 162 to 280°C
Set temperature:    500°C   Actual: oscillates between 242 and 321°C

Twiddling the airflow knob resulted in similar sh1tty numbers.  It appears it has a bang-bang controller, but with ridiculous hysterisis.   Oscillation period is about 10-15 seconds, with MAX and MINs on each oscillation being all over the place too.

1. Has anyone measured their ZD-xxx
2. Has anyone FIXED the temperature controller on their ZD-xxx

[frogblender]

Update to POS known as ZD-939A:

Ripped it apart,  the heater and temperature probe are both in the handle (as noted by others).   Measured the millivolt temperature probe output, it lags the actual air temperature horribly.... as if they encased the temperature probe in the best insulation known to mankind....

[frogblender]

Update #3 to POS known as ZD-939A:

Well, I ripped apart the handle, and sure enough, the thermocouple was completely encased in masonry/refractory/cement-ish stuff - it was manufactured like this.  I chipped off some of the cement at the tip, and a thermocouple peeked out.

Put er back together - Now, the air temperature is totally, completely, 100% kok-steady rock-solid  - ±1°C (the resolution of my meter) - beats the hell out of the ±60°C as shipped from the factory.   Instead of 15 second bangbang time, the heater light now flickers 2 or 10 times per second (which may lead to other problems with the SCR/triac/whatever that switches the half-rectified AC mains to the heater.... poor analog design with no hysteresis built in, evidently).


If I ever meet the boneheads who: 

a) designed/built this thing with such a dumbass rookie error, and,
b) shipped it without bothering to test if it bloody worked...

the whole lot of them will get a swift kick in the nads.

Bah.

[frogblender]

Forgot one thing:   this heap'o'crap has absolutely NO THERMAL SAFETY SWITCH WHATSOEVER - if the air pump fails with the heater on, there WILL be flames.

And it has one of those dumb-ish cooldown things, where you flick the mechanical power switch OFF, and it'll run the air pump full blast for a few minutes.  Trouble is, you get used to the air hissing sound, whether it is ON or OFF, and learn to ignore it.
A couple of times I've come into work in the morning to find some dork (possibly me) left the bloody thing on all night.

Bah.

[Wasp]

I know I'm late to the party. Got this old device practical new because it didn't and does not heat correctly. Although the display claims 400 °C it doesn't heat over 60 °C presumably. (Have nothing else to measure besides my hand.  )

Suspecting the heating element to have a problem, at least that's one of the few thinks *I* can check confidently. Before I take the handle appart and maybe be proven wrong: does anybody know what the resistence of the two circuits is resp. should be?
(I Meassured 2.2 Ω for the "white" pair and 106 Ω for the "black" pair.)

Out of curiosity: does any body know why there are 2 circuits for the heating element? Stupid of me. One must be the thermocouple.