Weller WE1010 Stopped working and troubleshoot

[shkhamd]

Hi anyone using weller WE1010 soldering station?
I have this unit (my mains is 220VAC 50Hz so I use it with a 220v-110v step down converter) and suddenly it stopped working.

So I opened it up and did the following tests:

1. transformer primary side shows 8 Ohm, secondary side shows 0.5 Ohm resistance

2. Secondary Polyswitch and Fuse was good.

3. At power applied, primary side has 100VAC 50Hz, secondary side shows 19VAC rms

4. The onboard fuse on the PCB was also fine.

5. But the 470 marked resistor ( R22, is it a 3 watt one?) on the attached picture shows a variable value roughly from 250 KOhm - 1M Ohm.

6. The rectifier diodes(D4, D3) were good and forward voltage was 0.6V tested with my DMM.

7. But the rectified DC voltage across the electrolytic cap showing around 2-3V, so obviously the 7805 voltage regulator was not supplying any DC current to the main controller.

I have tried to reconstruct the rectifier circuit from the PCB traces.
Please see the schematic picture I have added.
I don't recognize either of the diode markings (729 ZHK and A4) , nor I found anything useful by searching them on google. there ar lot of diodes with similar markings.

Also the diode rectifier part is a bit odd to me. Is it possible that D4 is a zener dropper in combination with R22 and the D3 is only acting as half rectifier?
In that case R22 is damaged by an overcurrent (may be an unusual in rush or surge) condition makes sense to me ( just suspecting, I am not an expert on electronics).
Also I don't understand why didn't the fuse tripped first in that case.

So after all these tests my guess is the 470 marked resistor is somehow gone bad.
But I am still confused to replace it with a good 47 Ohm with it because I m not certain about the root cause that made the resistor to fail.

Any ideas anyone?

[montarbo]

Hi,
R22 in that package can in no way handle 3w. Change r22 with TH resistor and test if u r not getting any short to GND beyond the resistor.

[Armadillo]

D4 is a 5.1V Zener part number SMAZ5V1.

Edit; [Some Cal Error]


The Design Current is 0.1A;

The resistor heat up and failed with time. Edit The Voltage Drop across the Resistor is 5V at 0.53W

[shkhamd]

I m deeply disappointed with weller, their design choices are so bad, how many cents they saved from not using a full bridge rectifier ic with a big TO220 package of 7805 ( 2 parts instead of 4 parts).

The total current of the microcontroller, LCD and MOC opto should not cross 100mA in ideal condition, since the controller circuit is just a specialized dimmer circuit with a BTA12 triac.


Anyway, I dont have 3watt zeners but I do have 1watts.

Can I use 3 1watt zeners in parallel?

One more thing, The zener voltage is a bit confusing.
Why it has to be a 5.1V zener? 19VAC rms becomes around 25V at peak, so a drop of 5volt is still 20volt acrosa the 7805.

So to drop 15 volt on the 7805 input the dissipation at say 200mA becomes 3watt.


Sent from my Nokia 7 plus using Tapatalk

[wraper]

Hi,
R22 in that package can in no way handle 3w. Change r22 with TH resistor and test if u r not getting any short to GND beyond the resistor.

Why would it need to handle 3W?

[wraper]

Current consumption should be tiny, well below 50mA. And 47uF capacitor with a half bridge rectifier is a good indicator for that. There are no signs of overheating on resistor itself or PCB around it. Most likely it failed due to mechanical reasons. Largish SMT parts are sensitive mechanical stress. Also it might be just a crack in solder joint. Picture is not close enough to see how solder joints look like, though.

[wraper]

I m deeply disappointed with weller, their design choices are so bad, how many cents they saved from not using a full bridge rectifier ic with a big TO220 package of 7805 ( 2 parts instead of 4 parts).

Using full bridge rectifier won't add any reliability. There is no need for vreg in TO-220.

[Armadillo]

I think your comment was a result of my earlier wrong cal which has been corrected. Please read the updated one above. Thanks


Quote from: shkhamd on Today at 05:08:07 am

I m deeply disappointed with weller, their design choices are so bad, how many cents they saved from not using a full bridge rectifier ic with a big TO220 package of 7805 ( 2 parts instead of 4 parts).

The total current of the microcontroller, LCD and MOC opto should not cross 100mA in ideal condition, since the controller circuit is just a specialized dimmer circuit with a BTA12 triac.


Anyway, I dont have 3watt zeners but I do have 1watts.

Can I use 3 1watt zeners in parallel?

One more thing, The zener voltage is a bit confusing.
Why it has to be a 5.1V zener? 19VAC rms becomes around 25V at peak, so a drop of 5volt is still 20volt acrosa the 7805.

So to drop 15 volt on the 7805 input the dissipation at say 200mA becomes 3watt.


Sent from my Nokia 7 plus using Tapatalk

[shkhamd]

I think your comment was a result of my earlier wrong cal which has been corrected. Please read the updated one above. Thanks


Quote from: shkhamd on Today at 05:08:07 am

I m deeply disappointed with weller, their design choices are so bad, how many cents they saved from not using a full bridge rectifier ic with a big TO220 package of 7805 ( 2 parts instead of 4 parts).

The total current of the microcontroller, LCD and MOC opto should not cross 100mA in ideal condition, since the controller circuit is just a specialized dimmer circuit with a BTA12 triac.


Anyway, I dont have 3watt zeners but I do have 1watts.

Can I use 3 1watt zeners in parallel?

One more thing, The zener voltage is a bit confusing.
Why it has to be a 5.1V zener? 19VAC rms becomes around 25V at peak, so a drop of 5volt is still 20volt acrosa the 7805.

So to drop 15 volt on the 7805 input the dissipation at say 200mA becomes 3watt.


Sent from my Nokia 7 plus using Tapatalk

yes, didn't see the updated reply [emoji56]

Sent from my Nokia 7 plus using Tapatalk

[shkhamd]

I m deeply disappointed with weller, their design choices are so bad, how many cents they saved from not using a full bridge rectifier ic with a big TO220 package of 7805 ( 2 parts instead of 4 parts).

Using full bridge rectifier would only increase dissipated heat in resistor and zener and won't add any reliability. There is no need for vreg in TO-220.

actually I m curious why it needs a zener dropper at the first place?
Can't a sub 100mA current draw easily be handled with the 7805 from 25V rectified DC?

Sent from my Nokia 7 plus using Tapatalk

[wraper]

Can't a sub 100mA current draw easily be handled with the 7805 from 25V rectified DC?

They just distributed heat over several parts. And there will be more than 25VDC when powered from 120V AC, getting very close to 35V absolute max rating and not having any room for increased mains voltage.

[Armadillo]

The 7805 will heats up. In Good Design, it is +3V to limit voltage drop across 7805. So 5V + 3V so that the 7805 will operate properly.
You may consider the LDO type where the +3V can be reduced much more, say + 1.2V.


Quote from: shkhamd on Today at 05:38:15 am

actually I m curious why it needs a zener dropper at the first place?
Can't a sub 100mA current draw easily be handled with the 7805 from 25V rectified DC?

Sent from my Nokia 7 plus using Tapatalk

[KL27x]

5V 100mA from a 25V input using a linear regulator is 2W. It would be silly to dissipate that much heat in a TO-220 part, IMO. If you want to use a linear regulator, get rid of some of that voltage drop in a power resistor. This is much more reliable than sticking heat sinks on the regulator and crossing your fingers. This is like taking a honda civic and putting in a 10.0 liter V8, but using it to generate 90% heat and 10% power.

I've stuck a D2PAK 7805 on a 20V rail to power 20mA. Just an indicator LED and a microcontroller, so I figured it would be ok to be lazy. This was plugged in 24/7. It worked a couple of years before it died. Come to think of it, I also had a heatsinked TO-220 7805 on a 20V rail powering just a panel meters and a couple micros. When not in use, the panel meter was turned off, so it was really just a couple micros on there sipping power. There were a few holes drilled in the enclosure over the heatsink, but no fan. The regulator died in about a year or two. As I recall, it died at least twice, which is why I added the heatsink and the holes.. and eventually learned to use a series power resistor. 2W is nothing to a resistor that is happy to continue working at 70C, indefinitely. It is a lot to an IC.

I vaguely recall someone trying to explain to me why this is bad, like 12 years back, at the time I did such things. That dropping 10's of volts in a linear regulator is not ideal, and the regulator might flake out years down the road. Well, I had to learn the hard way, but at least only on one-off stuff that was easily fixed.

Wraper is on point on all his posts. FWIW, the Hakko 888 also uses a half-bridge for the control circuitry.
 

[floobydust]

It's the possible high mains voltage due to the isolation transformer, when the heater is off, the voltage would go above 120V and overloaded R22 or C2 is damaged. But Dave did test that... 
R22 and C2 do see very high peak currents.

[shkhamd]

Okay guys, new update.

I have replaced the R22 resistor with a 3watt 47Ohm one, the D4 zener diode with a 1N4733.

Its now working fine, for at least the few minutes I have kept it under observation. I will have to wait for a longer stress test until next week.

Now my observation is I measure 116VAC rms at primary,
25VAC rms at secondary

but 44VDC across the IN and GND of the 7805!!!
The drop across the zener is 1V in DC mode of my DMM!!!
the drop across the 47 Ohm resistor is also less than 1V in DC mode of my DMM!!!

But the output at the 7805 is as usual, 5V DC.

I am not sure what's happening there but the the display and temperature controller is now working as before!

Is it possible that I measured them in some sort of wrong way?

Sent from my Nokia 7 plus using Tapatalk

[wraper]

I have replaced the R22 resistor with a 3watt 47Ohm one,

The drop across the zener is 1V in DC mode of my DMM!!!

That means current is tiny, around 20mA, just as I said before. Power dissipation is only 20mW. As I said resistor failed not because of overheating. And the only thing 3W resistor would do is rip the pads off  .

the D4 zener diode with a 1N4733.

Why would you do this? Likely you replaced it with zener rated for wrong voltage. And likely placed it in wrong direction on top of that.  There shouldn't be 44 VDC so probably you made measurement wrong.
All you have done overall is significantly reduced reliability.

[Armadillo]

Better Diode Mode Check D3 whether its damaged e.g. shorted.

[wraper]

BTW all of those who suggested modifications without even knowing actual current consumption and heat dissipation. IMO you should be ashamed for suggesting something that can lead to more damage without any valid reason.

[Armadillo]

The only possible cause of your weird measurement result is D3 is not rectifying.
You should measure it out of circuit to be sure, if in circuit measurement confused you.
Or just replace D3 for double assurance.

BTW, you sure your modification to higher wattage, you can close the enclosure?

[wraper]

The only possible cause of your weird measurement result is D3 is not rectifying.
You should measure it out of circuit to be sure, if in circuit measurement confused you.
Or just replace D3 for double assurance.

BTW, you sure your modification to higher wattage, you can close the enclosure?

Just put everything as it was originally and don't do any further nonsense.

[floobydust]

L78M05 is rated max. 35V in  so measuring 44VDC is bad news. Is the new zener shorted or what is the transformer secondary voltage.

DMM measurements can lead you on goose chase because the voltage is DC with 1/2 wave ripple.
Also you need true RMS to measure ACV because there is lots of sine-wave distortion with cheap transformers. It's rated 23VAC under (heater) load and 31VAC is what you are getting at no load (to maske 44VDC) but you measure 25VAC?

The entire board is low power, I would guess 25mA with opto on. You can go through the math, but the Vreg must get much lower DC input.

[rsjsouza]

I can't help but wonder if the Zener was mounted in reverse, where it would simply drop its 0,6~0,7V instead of the rated 5,1V.

[shkhamd]

Okay rewinding what I have done so far:

1. I have replaced the bad 47 Ohm resistor which I believe is obvious, may be overkill but should not hurt the circuit.

2. I have replaced the D4 with a 1N4733, things that I could do wrong with it is, place in reverse polarity or use a wrong voltage value.

Well, I have so far triple checked its polarity for being in correct orientation. I measured the forward voltage in diode mode and made sure that both the original and replaced diode shows the Vf from the same side of the PCB pads. Only difference is the diode original had Vf about 0.6 and 1N4733 has Vf about 0.7.

I don't know how to check for the zener voltage.

With the original SMD zener diode, the black terminal of the zener placed now was the pad where the cathode was and the other side was the anode.
So in my present replacement it is still the cathode as you can see in the previous pictures.

Shouldn't the positive voltage be applied at the cathode to make it act as zener, not sure if my understanding of zeners is wrong.

Now is it possible that a zener diode may be damaged in such way that it will show it Vf just as a normal diode but when placed in circuit in zener configuration, it will act as a short?

3. I also did checked the Vf of D3, the rectifying diode both in forward and reverse bias. Vf was 0.6V and in reverse bias it showed nothing or out of range for as if it was a very large value.
So i believe D3 is also good.


Edit: And I also powered the board up once and first I replace the resistor with the original diode still in it, it worked but I didn't take any measurement of the voltages though. Reason I replaced it later because 1N4733 is rated at 1watt.

[shkhamd]

Yes I also checked first if the housing can contain the resistor with in it and fortunately there was quite enough space under the front cover.



Sent from my Nokia 7 plus using Tapatalk

[Armadillo]

Take a meter, check the trace from the Resistor, which pad it goes to, then that pad should be connected to the cathode.
The picture is not clear to see the trace from here.