-
I posted a question regarding the Triac in the attached circuit in another message, but that was regarding something else, this is regarding a repair (just want to keep myself honest so nobody thinks I'm double posting).
My WES51 Weller Iron stopped working without warning - the LEDs didn't turn on nor did the iron heat. Yesterday I disassembled it and took some measurements. Fast forwarding to today (reference attached schematic) I confirmed the output of the transformer is 24VAC and 5V is coming out of the LM78.
At first I thought (and still remain a bit confused over what I measured) that the thermocouple inside the iron itself opened, however, looking at the circuit, I realized I can unplug the iron and check if the op-amp is toggling when I pull down (or leave high) the bottom side of the 1M resistor (essentially pin 3 of the op-amp).
Upon doing this, I managed to see that the op-amp was either 0V or 5V, and, I could see the green LED go off and on when I either pull up or pull down (I forget which one caused the LED to turn green). Unfortunately in neither state did the red LED illuminate.
From looking at the circuit, I believe the iron can remain unplugged and not have an affect on the Triac section.
Having said all this, the op-amp is toggling, I have 5V, 24V, and the Triac doesn't provide any feedback to the PIC to tell if the iron is actually heating thus allowing me to leave it unplugged (to rule out external anomalies).
The one pin I'm uncertain of: pin 5 of the PIC is only 3VAC. My feeling is that's an RC circuit and it should be 24VAC, however, maybe the PIC has low impedance and loading down the voltage.
The major issue that I see, and I'm praying it's not the PIC: when I toggle the op-amp, the base of BC547 isn't going high or low; it's only going to about 34mV.
From all I measured, I've almost determined the PIC is bad, but hoping pin 5 should be at 24VAC and that maybe the capacitor is resistive (obviously I can remove it to determine this).
On a side note, I have a junk backup iron, so I am limited on how much I experiment with removing components.
-
Pin 5 should be a line frequency near-squarewave, clamped by the chip's internal input protection diodes, with the bottom a bit below Vss and the top a bit above Vdd, but not by much as the 1Meg resistor only allows a peak current of about +30uA, -34uA (asymmetric as the Vdd and Vss rails aren't centered on AC 0V) through the protection diodes. Expect the 'squarewave' to be about 6V pk-pk, approx 0.5ms risetime and to have slightly humped top and bottom due to the increased protection diode Vf at peak clamping current.
Try disconnecting R14, and see if the PIC output on pin 2 is any better when unloaded. Also check the pot can sweep the setpoint voltage on pin 6 from 0V to 5V. -
Quote
Try disconnecting R14, and see if the PIC output on pin 2 is any better when unloaded. Also check the pot can sweep the setpoint voltage on pin 6 from 0V to 5V.
I didn't necessarily try sweeping the pot, but I manually set it to the middle (I had to remove the knob to work on it) and I'm measuring almost 2.5V at pin 6; but I'll sweep it.
I'll place a scope probe on pin 5 and hope the PIC is fine. Due to bypassing the soldering iron thermistor, it seems a safe bet the iron isn't the (main) culprit.
Good idea about removing the resistor. I should have checked the transistor, but thought at 47k, a shorted transistor probably wouldn't show near 0v on the PIC side. -
To help you locate the fault, consider the following...
a. K-type thermocouple produces 42µV/°C;
therefore its output is: 8.4mV@200°C, 12.6mV@300°C, 14.7mV@350°C, 16.1mV@384°C
b. the opamp non-inverting configuration has a gain of about x310 (1+464k/1k5, but is likely lower due to 10R/14K divider);
therefore its output is: 2.604V@200°C, 3.906V@300°C, 4.557V@350°C, 4.997V@384°C
Q: why did I specify 4.997V@384°C?
A: because that's probably what the ADC uses as a reference and can't be >Vcc.
c. pin 5 of the MCU probably senses the zero-crossing of the 24VAC secondary.
d. an unconnected thermocouple should cause the opamp to output +5V (or close too it) which is equivalent to "too much heat" and therefore the MCU should shutoff the triac.
e. a shorted thermocouple should cause the opamp to output 0V (or close too it) which is equivalent to "too little heat" and therefore the MCU should turn on the triac.
f. when transistor base is toggled, the 2.2µF cap on the triac gate will cause the triac to conduct for only a 1/2 (24VAC) cycle before needing to be retriggered again so just grounding the base of the BC547 won't be able to test that part of the circuit. The MCU is very likely sending a train of pulses and that's why you are measuring 34mV (avg) here.
Q: why is there a cap anyway?
A: probably a safety measure; just in case the triac gets stuck-on.
Q: Where is the reed switch located? Given the station's design, It doesn't serve the usual purpose of putting the station in standby mode. So what's the purpose of it? Are you accidentally triggering it and therefore messing with your fault finding?
I don't know the meaning of red/green LED but the pin can output either 5V (red on), 0V (green on), or hi-Z (set to input mode so that neither is lit).
I would expect in a working station that if you were to place a resistor (say, 470R or 1K 1W) across the heater connector contacts then short the thermocouple contacts, you should should measure 24VAC across the resistor.
The iron heater contacts should measure 10 ohms.
P.S. I don't own a WES51 but I do own a PES51 (the iron) and have built a solder station of my own design around it.
-
Quote
Q: Where is the reed switch located? Given the station's design, It doesn't serve the usual purpose of putting the station in standby mode. So what's the purpose of it? Are you accidentally triggering it and therefore messing with your fault finding?
I don't know the meaning of red/green LED but the pin can output either 5V (red on), 0V (green on), or hi-Z (set to input mode so that neither is lit).
Wow, thanks for the detailed math explanation. The reed switch (unless I'm wrong) is a silly way of locking the temperature. This iron has an optional 'pen' someone can purchase which is nothing more than a magnet. The manual states the steps to lock the temperature. I'm assuming if this were on a manufacturing floor, and parts couldn't be soldered any higher than 550 degrees F, then a manufacturing engineer could 'lock' the iron to that specific temp and moving the dial won't change it. The manual explains how to reset the reed switch so the temp can manually be set again, however, I don't know anything about PICs, so I'm uncertain if setting the temp (via the reed switch) actually commits this to memory or if the unit can be re-powered (without the magnet being close to the reed switch).
I'll admit, initially I didn't know what the reed switch was. I thought it was a secondary fuse and it opened. After digging a bit deeper, I quickly got the answer, and, once referenced the manual, realized the switch is exactly where the magnet needs to be placed.
The red/green LED is now a bit confusing since I never needed to commit the sequence to memory. Now that the unit doesn't work, I'm kicking myself for not remembering. I think it remains red as it's approaching temp, and blinks green when it has reached temp, but both off if temp is decreased. It's a two-in-one (?) LED with green and red in one package.
When I measured the pins of the iron itself (unplugged from the base), the thermocouple was absolutely confusing me. It was several mega ohms and seemed to increase indicating the meter was causing a charge or whatever. I grabbed a hair dryer and heated the iron as I measured the thermcouple, it seemed to decrease, but then suddenly began climbing again. Needless to say, it was all over the place and I expected it to be 100k ohms instead of several mega ohms (and inconsistently moving around in resistance by a great amount).
I decided to bypass the iron if possible and trigger the thermocouple pins on the base manually thinking I could tell the PIC the iron was calling for more/less heat. As mentioned, the op-amp is working fine, but, as you mentioned, if the PIC isn't detecting some cross point (such as some sort of feedback I assume?) then it's not 'pulsing' the gate.
Maybe the station is good and the thermistor in the iron itself is bad? I have 100k thermistors I believe and can try connecting them to the station, however, I'm uncertain what the value in the iron really is.
Thankfully this circuit doesn't have much to it, but unfortunately, if it's the PIC or the iron, I'm somewhat screwed. -
Divide and conquer.
Always check power first; 5VDC and 24VAC at the correct points including the chip power pins. Although, I think you've confirmed that already.
Connect the iron to the station, measure pin6 of the op amp (multimeter on DCV). Remove the tip from the iron to expose the thermocouple and use a hair dryer to heat the open end up. Pin6 should go from 0V (or close to it) to multiple volts (as per my calculations) depending on how much heat you apply. If you let the iron cool, it should go back to the starting point. Confirm that this same voltage is on pin7 of the MCU. If this is what you see, then you can conclude that the thermocouple, connections, opamp is all good. If not, work backwards.
Typically, you won't be able to measure the thermocouple output directly with a hand-held 3 1/2 digit multimeter because its lowest range starts at 0.0 mV. My 5 1/2 digit bench meter can directly measure it as its lowest range starts at 0.0µV. But you may see some activity in the low mV but I don't think the hair dryer will inject much heat. And don't use a candle (too much soot and/or it might bake it).
I too found it difficult to measure the resistance of the thermocouple; couldn't get a firm connection free-hand which is very imporant given the low voltages involved. Although, its definitely a K-type thermocouple and not a thermistor.
If you want the test the station without the iron, execute the procedure in my last message.... short the thermocouple pins in the connector and measure the ACV across a 1K resistor connected to the heater pins. The green LED should be solid. If you have an oscope, there should be a train of pulses from pin2 of the MCU and you should also see pulses on pin5 of the MCU coming from the zero-crossing points in the AC wave of the transformer secondary. This should happen even on the lowest temp setting.
The key here is that there will be a steady voltage from the opamp between 0V..5V into the MCU. To confirm that the MCU is working, you could go so far as to disconnect the leg of the 14k resistor (opamp output side), attach a spare 10K pot to 0V and +5V (end pins, respectively), and wiper pin attached to the loose 14k resistor leg. Then you can adjust this pot as a substitute for the temperature being sensed. ie. 2.6V on wiper corresponds to 200°C, 3.9V corresponds to 300°C, etc.
If the iron is dead (bad heater or thermocouple) then a replacement is about $50.
If the MCU is truely dead and if you're up to it, you could always substitute your own with your own programming.
-
First I would repair the wand. Inside is a thermocouple, it should not be shorted to anything (i.e. heater) nor open-circuit.
Weller does have some fault checking in their firmware. If the handle is unplugged, it detects the open thermocouple and might go to lockout, where you have to cycle power to get the station running. So playing with getting the op-amp to output 0V and 5V might go to safety lockout.
So I wouldn't blame the controller just yet, it is not turning on the heater because it thinks things are unsafe.
You did replace the two electrolytic capacitors? The 33uF part is highly stressed and does not last.
The WES51D eevblog user Emrtech reverse-engineered and posted schematic and scope waveforms, same basic hardware but a larger PIC and LED display.
Weller did not code protect so you can read out the PIC firmware. WES206 controller board I think is obsolete. -
Quote
Divide and conquer.
Always check power first; 5VDC and 24VAC at the correct points including the chip power pins. Although, I think you've confirmed that already.
This is usually how I approach things. Typically I try removing any items that can be removed without necessarily affecting the circuit. Upon looking at the circuit, and before disassembling the iron, I thought this would be a slam dunk repair. Once I saw a PIC, I sensed it could be trouble.
Also, on a side note, I'm fully aware this is a thermocouple, however, after many, many years of working with thermistors, I constantly call them thermistors by accident; so bare with me.
One of my temp meters (four channel) has K-type thermocouples that I figured could be used to trick the base. Upon powering it, the green LED illuminated, so I began scope probing pin 6 of the PIC, and suddenly the LED turned red. It almost seemed like each time I touched the probe to the pin, the LED would change colors, however, then it began doing so on its own. I wiggled the power wire from the transformer, and it seemed to consistently cause the LED to change so I used my (cheap) soldering iron to reflow the two power wires.
After I connected the iron which still remained cold, and that's when I probed the resistor between the base and PIC. It had 250us wide pulses occurring approximately every 8ms. I asked myself how its getting pulses, yet the iron remains cold, so I touched the iron to double check, and burned my finger.
Now things are still screwed up because if I apply power to a cold iron, the dial on minimum temp, the green LED is on (I think it should be red), and the green LED never seems to turn red or off.
Not knowing the temperature of the soldering iron (which was enough to melt solder, and burn my finger), I connected one channel of my four channel temperature meter to the iron (obviously I didn't have the best thermal conduction, but it was enough for a rough idea) and saw the iron settled to about 500 degrees F.
Regardless of where I placed the pot, the thing seemed to have a mind of its own. Sometimes it would eventually begin to cool (long after I turned down the pot), or other times it didn't begin heating immediately, but in both cases, it seems the LEDs weren't showing the correct state.
I need to somewhat start my analysis over, and check the 33uF cap. Maybe it's causing lots of noise (or one of the other caps), but I measured pin 5 of the PIC had about a 6v peak 60Hz wave that looked fairly clean.
My only other thought due to these strange occurrences: the PIC is flaky thus why it's not triggering the correct LED state and not getting correct thermocouple feedback.
Edit: reading the manual (who reads manuals nowadays?) maybe the red LED was never on. From my interpretation (and wanted to edit this quickly before someone spends time trying to analyze why the red LED isn't on) it looks like red is only for 'locked' mode.
A steady green is the temp ramping up, blinking means it has reached temp. -
Some good news and bad news, and also some technical feedback for anyone repairing their WES51.
First thing I dug deeper in was checking the ripple on the input of the 5v regulator. With the scope on AC coupling, I saw 7V p-p which seems quite excessive and believe the 33uF cap is bad. Unfortunately, the 5V on the output only had a ripple of about 100mV p-p. I was hoping to see ripple on the 5V also indicating the circuit is too noisy.
The bad news about this iron: it's now somewhat working.
A more consistent observation: upon powering, the iron will not heat. If I touch the scope probe (which has the ground lead connected to the iron ground on the secondary of the transformer) to R14 (transistor base resistor), the iron begins heating. It seems to heat to approximately what I have the dial set to (minimum setting which is about 325 degrees F), and will somewhat settle. If I turn the knob to a higher temperature setting, the iron heats further and seems to settle. If I turn down the knob to a lower temperature setting, the iron cools.
The issue: the green light stays on constantly. After reviewing the manual and actually seeing a somewhat working iron, my mind began remember the LED status: the red LED has never come on (although in my previous posts the red LED had been coming on which confused me - I don't know why it was turning on since it technically shouldn't - unless the PIC is going bonkers). The normal LED function: the green will stay on constantly while its heating and then blink when it reaches temperature. I think if I decrease the temperature, the green LED will turn off until it reaches the lower setting.
I'm not seeing this now. The green light (as I stated above) isn't turning off and remains on in all states.
I'm seeing about 1v on the output of the op-amp when the temperature is about 320 degrees F (160 degrees c), however, this is a sine wave that seems to jump around. R14 continues to have 250us pulses occurring every 8ms, but will stop pulsing if I decrease the temperature and the iron needs to cool in order to reach the lower setting.
I accidentally touched pin three to pin four on the PIC, and this caused (what I believe) a lockout because the red LED came on. I managed to turn off the lockout by applying a magnet to the Reed switch, and the green LED came on. To see how the PIC reacted, I played with the lockout feature a few times. The red LED would come on, green would come on after unlocking it, etc...
Trying to figure out if the iron is holding temperature, and the correct temperature, is hard. I have the tip sitting in a piece of copper tape with the thermocouple (that goes to a temperature meter) also taped to it. It's a small piece of copper tape, but there is lots of room for error due to air gaps.
Sometimes when I recycle power, I have to touch the scope probe to R14 again in order for the iron to heat (but not always), but the green LED doesn't seem to change state (it's always on).
From reading other posts about the other model, someone mentioned a bad ground. I'm wondering if one of the wires isn't grounded well causing the PIC to not set the correct state of the LED, or if the soldering iron temp is bouncing too much due to a bad connection and the PIC isn't ready to set the green LED to ready.
-
If the LED is solid green but heating doesn't occur until taping with the 'scope probe... hmmm...
Isn't one leg of the transformer secondary (GND of LM78L05) bonded to the case/3rd prong of mains plug?
Again, clip your multimeter (set to ACV) to the heater connector pins and try to reproduce. If you don't see 24VAC until you tap it with the 'scope probe, then it may be that the triac isn't getting enough current to trigger. Check for dry/cracked solder traces around the, transistor, triac, connector, or change the 2.2uF cap on the triac gate to same or 4.7uF, or change the triac.
-
I would look for a ground-fault because the scope is earth-grounded and so is the soldering iron wand (heater/tip). Connecting a scope probe should not have side effects, it's a 1MEG//xxpF load to earth-ground and should not trigger a triac? Is that 2.2uF cap good?
Check continuity from the soldering station's third prong to the iron's tip, as well as the secondary-side/controller board. Any chance the tip touched some high energy circuit and a ground-fault melted a wire/pcb trace inside the station?
If the heater shorts to the tube, I'd expect it to blow the fuse or roast the triac.
If the thermocouple shorts to the tube there would be strange temperature readings.
If the heater shorts to the thermocouple, the polyfuse would trip. -
Just a FYI, I measured my PES51 iron with a 6 1/2 digit bench meter...
Given:
- tip and shroud off (just to make sure it wasn't the tip that shorts thermocouple to metal body),
- negative lead clipped to metal body threads just below the insulated handle,
- meter on 2-wire mode with positive lead nulled to the threads next to the negative clip.
Positive lead on the GND/Shield connector pin: 0.20 ohm
Positive lead on the thermocouple negative connector pin: 1.10 ohm
Positive lead on the thermocouple positive connector pin: 4.60 ohm
Neither heater pin is connected with the metal body. However, there's 10 ohms between heater pins.
It would seem pointless to have a fuse on the thermocouple negative pin to GND given the 0.9 ohm short to the GND/Shield pin.
-
Touching the base of the transistor resistor to get the iron to heat doesn't always seem to be necessary.
Things just seem weird and confusion is occurring due to having a few unknowns: what the PIC is suppose to do and the actual iron temp (I can only assume I'm getting good thermal transfer between the copper tape the iron is touching and the theromocouple connected to my external meter).
Anyway, as mentioned, the first thing I noticed was 7V p-p ripple on the input of the voltage regulator. While this seems excessive, and most likely the filter cap is bad, it should be enough margin because it's 24V AC down to 5V DC. What I noticed yesterday was very weird:
The 5V DC periodically goes from a steady state ripple (think I had said 100mV p-p) to a sloppy looking sine wave while the iron is heating. Also, the output of the op-amp sporadically goes from DC to a 1V peak perfectly formed sine wave for short spurts. I removed the iron and measured voltage to ground at pin 3 of the op-amp. It was 4.6V DC which confused me because the op-amp should be high impedance and the thermocouple is not in the circuit. Thinking something is leaking to ground, I measured pin 3 to ground and saw 534k, the 1M was around 536k, and 5V to ground was about 359ohms.
I thought maybe the 1M was damaged (I've actually seen a surface mount resistor change value - maybe the end cap was damaged and caused the resistance to change), but out-of-circuit it measured 1M. I also removed the capacitor from pin 3 to ground and that didn't change the voltage at pin 3.
The 5V line to ground having 359 ohms somewhat makes sense because the other circuits are voltage dividers to ground, but not being able to measure 1M with the iron unplugged seems odd, and having 4.6V DC on pin 3 of the op-amp seems odd too.
The PIC is definitely not changing the state of the green LED. The iron heats and seems to control (based on my temperature measurement setup) and the PIC doesn't tell the LED to blink (although I saw some random positive pulses here and there on the scope). If I increase turn the pot to increase heat, the iron heats further, and, when I turn down the knob, the LED remains on (I believe it's suppose to turn off until it cools to the new temp setting).
It almost seems like the voltage regulator is bad due to periodically oscillating, the op-amp is bad because it has resistance to ground causing 4.6V DC on the input of pin 3, and the PIC is bad because it's not blinking the LED. Yet, the iron heats and regulates which tells me all these anomalies don't matter and the PIC is the culprit because it's not blinking the LED (yet, it's working because it controls the heat).
-
I took some oscilloscope captures of various points (see attached zip file).
The file names are self explanatory. The one labeled 5V glitch is some glitches I kept seeing on the 5V DC line that I can't explain. Also, I'm uncertain if the pulses should be going below ground (although my scope probe ground is on the secondary that is circuit ground).
I can't tell if touching a point with the scope probe is what starts the heater or if it's coincidence. Yesterday it seemed once I connected the ground on the secondary of the transformer to the scope probe ground, I got heat. Tonight I tried touching ground to scope probe ground, and nothing happened. After a few minutes the iron began getting hot.
It seems the PIC is bad since the light doesn't blink, but yet, it also heats the iron and seems to settle to the set temperature.
-
The scope waveforms look fine to me. The 5V rail glitch is 150mV and 60nsec it's more likely an artifact from the scope and grounding. The last trace Top of Triac.pdf I can't make sense of.
Did you check continuity with the station's ground and the control board, the third green wire, when there's no scope connected?
The soldering station is acting like the thermocouple is open when cold and AC mains hum there gets it to start and once hot the thermocouple sort of works and is no longer an open circuit.
Does the red LED+PIC work? It's red only for "unlocked mode" when the reed switch is activated. The green LED always on, slim chance could be one bad output pin on the PIC, I would put aside the notion the PIC is the problem. Maybe try another approach to troubleshooting this, like investigating what others have suggested.
Analysis can only take troubleshooting so far. -
I'll read through this thread again, but have checked things others have recommended.
Today it occurred to me I may have a spare iron (not base) and was going to try a backup iron, however, when I searched in the area it would be in, I didn't see it.
Tomorrow I'm going to keep it powered for a few minutes to see if it starts without me touching anything. I had tried the red LED by locking the temperature using a magnetic next to the Reed switch and the red LED lit.
On a side note, I was thinking of removing the PIC and downloading the code so I can post it for others; including having it for myself just in case I ever need it. Not sure if this is easy, but I have a programmer; unfortunately little experience with programming PICs.
-
Quote
The last trace Top of Triac.pdf I can't make sense of.
I accidentally didn't acknowledge this. That scope capture was taken at a point where it went haywire to show some of the stuff I believe is weird.
The only part I didn't understand is why things were dipping below 0V DC. I know the ground is referenced to the secondary which is +/- 24V, but, as an example, the collector of the transistor is 0 to 5V. When the iron is connected, it looks like the same wave, however, shifted below 0V DC by 4V DC. If it was referenced to ground via the secondary, then wouldn't it dip much lower, and, why does it dip when the iron is connected?
Also, I forgot to mention, to reduce ripple on the input of the regulator, I replaced the capacitor with a 100uF 25V - it's absolutely underrated voltage wise, but it's all I had. Figured I could try reducing the ripping to see if it made a difference, but plan to replace it. Since 100uF made such a big difference, I may look into keeping that value (obviously buying one with higher voltage).
-
I keep thinking there is a mechanical problem, which could be an open (when cold) thermocouple. A heater to thermocouple short would trip the polyfuse.
The PIC output pin is fine if it can light the red LED. I didn't see anyone read out the WES206 firmware and post on the Internet.
A triac needs gate current in both directions - so Weller is using the charge on the 2.2uF cap to supply negative gate current (when the drive transistor is on, until the cap discharges). With the heater disconnected, triac BTA12-600 MT2 is not connected and that -ve gate current has no path I think.
You can substitute a resistor or light bulb for the heating element, and a mV potentiometer spoofing a thermocouple to really test out operation without the handle, if you want to analyze more. But sometimes we can get fixated on the wrong problem and it's a lost repair where patience runs out. Treat it as learning more about electronics. These are very good soldering stations and new chinese crap terrible in comparison. Yuck.
The 33uF filter cap does see very high ripple current and does not have a long life. Weller went a bit cheap there. Even the new WE1010NA uses 47uF 50V and is lower power (LCD, no LED's). I would upsize to 100uF or so. -
I only had a few spare minutes tonight, so I left the iron powered for about a good 5min without it turning on.
Most likely I'll have more time Friday or over the weekend and plan to somewhat start from the basics and move on.QuoteBut sometimes we can get fixated on the wrong problem and it's a lost repair where patience runs out.
You're correct. I'll openly admit, I began this by separating things (i.e. measured the obvious 5V DC voltage) thinking it was the fuse, triac, or regulator. I dug too deep into the oddball signals I saw (5V glitches and a few other things).
Tonight I also played around with magnetizing the Reed switch. When I held the magnet long enough, the red LED would blink. After I held it again and the green LED would blink. So the PIC is capable of blinking the LED and maybe you're correct and this is purely mechanical.
Unfortunately a heater wire broke off the pin to the connector on the inside. I was able to solder it to make a connection, but, once I (hopefully) fix this, I'll need to get replacement pins (which hopefully I can figure out the part number and still be able to locate). -
Quote
Positive lead on the GND/Shield connector pin: 0.20 ohm
Positive lead on the thermocouple negative connector pin: 1.10 ohm
Positive lead on the thermocouple positive connector pin: 4.60 ohm
Neither heater pin is connected with the metal body. However, there's 10 ohms between heater pins.
Sadly I think it's the thermocouple because with the iron plugged in, I'm measuring approximately 541k across the orange and yellow wires. I'm uncertain the layout inside the iron, but I removed the tip and measured from the small pipe inside (which I assume is the thermocouple) to the PCB. From orange to tip, it was a few ohms, from yellow to tip, it was 524k.
What baffles me about the iron: if the inner tube which looks like it's copper is the thermocouple, once the tip is inserted, then it should be electrically shorted to Earth ground since the metal of the iron is Earth ground.
As the previous statement read, I focused too much on why the LED wasn't blinking, touching points with the scope probe seems to start the heating, etc...
I'm guessing the PIC has code that is sensing something is wrong, so it's not blinking, however, from taking temperature measurements, it seems the iron reaches the set point and the LED "should" blink.
Initially I tried using a k type thermocouple from a meter to trick the iron, but, if I remember correctly, that didn't seem to work. I thought maybe the resistance was lower than the resistance in the iron and caused the op-amp to think the iron is too hot.
Disassembling the iron doesn't seem possible, so all I can do is get a new iron.
Does any "k type" iron work?
Edit: looks like I can get the exact iron for $52 (model PES51) -
Quote
Positive lead on the GND/Shield connector pin: 0.20 ohm
Positive lead on the thermocouple negative connector pin: 1.10 ohm
Positive lead on the thermocouple positive connector pin: 4.60 ohm
Neither heater pin is connected with the metal body. However, there's 10 ohms between heater pins.
Sadly I think it's the thermocouple because with the iron plugged in, I'm measuring approximately 541k across the orange and yellow wires. I'm uncertain the layout inside the iron, but I removed the tip and measured from the small pipe inside (which I assume is the thermocouple) to the PCB. From orange to tip, it was a few ohms, from yellow to tip, it was 524k.
What baffles me about the iron: if the inner tube which looks like it's copper is the thermocouple, once the tip is inserted, then it should be electrically shorted to Earth ground since the metal of the iron is Earth ground.
There should not be that much resistance between thermocouple pins (if that's what the orange and yellow wires are). Thermocouples are just two dissimilar metals welded together. It's the difference in temperature between the pointy-end and the connector-end that generates a tiny voltage (41 µV/°C) that the opamp amplifies; not its resistance. https://en.wikipedia.org/wiki/Thermocouple#Type_K The bulk of the resistance (3.5 ohms) that I measured is probably just the thin wire from that center tube tucking inside the tip to the pins on the connector. That tube could just be a protective shell with the actual thermocouple wires suspended inside in some compound the same way heating elements are.
I was surprised (even without the tip) that the negative thermocouple pin somewhere inside the iron still has a connection to GND/Shield pin (0.9 ohm). It makes the 200mA fuse pointless.QuoteAs the previous statement read, I focused too much on why the LED wasn't blinking, touching points with the scope probe seems to start the heating, etc...
Flaky connection maybe?QuoteI'm guessing the PIC has code that is sensing something is wrong, so it's not blinking, however, from taking temperature measurements, it seems the iron reaches the set point and the LED "should" blink.
I believe that not blinking (solid green) means the temperature is too low; ie. starting from a cold iron so the heater should be on full-time. Only when it senses over-temperature does it start blinking (cutting off power to the heater); ie. hunting around the set-point.QuoteInitially I tried using a k type thermocoupe from a meter to trick the iron, but, if I remember correctly, that didn't seem to work. I thought maybe the resistance was lower than the resistance in the iron and caused the op-amp to think the iron is too hot.
Disassembling the iron doesn't seem possible, so all I can do is get a new iron.
Does any "k type" iron work?
Any k-type thermocouple will produce the same 41 µV/°C but just having it connected without it actually being warmed-up should just give a solid green LED. It's a feedback loop: cold TC ==> apply heater power ==> hot TC (past set-point) ==> cut power to heater (let tip cool) ==> still hot TC (but below set-point) ==> apply heater power ==> go to step 3.
But not any iron will do. It will need a K-type thermocouple, 10 ohm heater, the correct connector and pinout.
It shouldn't be too hard to find a replacement: https://www.google.ca/search?q=pes51+replacement+iron
-
Quote
I believe that not blinking (solid green) means the temperature is too low; ie. starting from a cold iron so the heater should be on full-time. Only when it senses over-temperature does it start blinking (cutting off power to the heater); ie. hunting around the set-point.
You're correct, solid green should be full heat turn on. From what I can tell, I am able to get the iron hot and it seems to come up to temperature (about 330 degrees F). At that point I increase the temperature and it's well over 500 degrees F, turn down the temperature, and the green LED remains on.
This is why I've wondered if something else is also wrong and began chasing other stuff thus making the silly mistake of ignoring the high thermocouple resistance. -
I've been reluctant to pull the trigger and spend the money on a new iron until I can figure out if the PIC is good.
Saturday I soldered a k-type thermocouple to the PCB and taped the temp end to a hot plate. I set the knob on the solder station to minimum and turned on the hot plate. The gate to the TRIAC was getting clocked, so it assumed it was heating a real iron.
Once the hot plate came up to the temp set on the station, the light remained on (it should have been blinking). I increased the hot plate temp and turned up the dial on the station so it thinks it's heating the iron more. Once it came up to about 375, I turned down the knob on the station.
Normally at this point (unless I'm wrong) the green LED will turn off until the iron cools to the new lower set temp. This didn't happen, the LED remained on.
After some research, others have found the issue where the heater turns off after connecting scope probe ground to the PCB ground, however, I haven't found a solution. I'm wondering if this station has two problems (the thermocouple in the iron and something else).
This is a weird problem. I'm going to try pulling the code off the PIC for the benefit of replacing it, but also so I can provide it to anyone else who wants it.
I contacted Weller and they said the iron is discontinued and any code is no longer available.
With this being said, I'm uncertain if my PIC is damaged (won't know until I try sometime this week), so thought to send the question in advance of whether anyone has the code, has an iron they don't care about and can pull the code, etc...
-
I have repaired a couple of Weller sets myself, but only the analog controlled ones so far.
Up to now, I had not a single problem with the station, except a damaged connector once.
I have only encountered these three failures on the solder pens yet: broken cable (at the pen side), broken sensor or broken heating.
Also intermittend sensor failure leading to premature functional assessment. -
Hopefully someone can benefit from my updates.
Tonight, with a k-type thermocouple soldered onto the PCB and the iron plugged in, the iron immediately heated upon powering without the need to touch the ground scope lead to PCB ground.
Others did say this sounds like an iron thermocouple issue from the beginning, and maybe they are correct. I can't figure out why connecting the iron causes some sort of ground connection when only five wires are used (two for the heater, two for the thermocouple, and one for Earth ground).
My initial thought was that Earth ground is nothing more than the metal of the iron. When the TRIAC is not conducting, that portion of the circuit is open. Maybe the PIC is detecting an iron heater is attached, but I don't see how the PIC can detect whether the iron is connected or not; unless Earth ground has some high impedance to the iron heater so it causes a voltage divider on pin 5 of the PIC.
In any case, I may gamble and order a new iron (unless someone looks at the code and think the PIC is corrupt - keep reading for an explanation on what I mean).
To perform one last test before ordering an iron, I removed the PIC and 'read' the code. Since it was able to read, I'm assuming the PIC program isn't corrupt.
I'm uploading the code in hopes others can benefit from this, however, I want to emphasize something AND ask a favor.
The emphasis is: I can't guarantee this code isn't corrupt. All I've done is remove the PIC, stick it in the programmer, and 'read' it. Obviously it can't contain a bug or anything, but I just want to emphasize that I don't know what this code means, however, I didn't do anything than save the buffer to the attached text file.
The favor: asking for a bit of assistance. I've never really 'read' PIC code, nor do I have experience with this programmer (I have used Xilinx to program FPGAs and other programmers to program pre-compiled HEX code for EPROMs. The software for this programmer only offered to save as a .txt file.
Is this text file all I need to program a new/blank PIC (the ASCII code doesn't look the same when compared to the buffer in the programmer software and assume the characters can't be deciphered as they can in the software so I fear this text file isn't the only thing I need)?
Also, can this code be interpreted into human readable words so we can figure out what is going on inside?