2021 Craftsman Bench Jointer CMEW020 - help me reverse engineer/hack this please

[jakeisprobably]

So the Craftsman CMEW020. I read the reviews on the big box store's website. They weren't good. I knew I would probably have issues, but the product warranty is 3 years, it was relatively cheap, and the only jointer available locally. Half of my reason for buying it was to post here if (after) problems... It's a Maker hobby-in-hobby-inception...
  It worked great for around 2 hours total before it died. I ran some 2x4's at the smallest depth of cut and at the easy end of motor speed settings (@ 2 of 5). There was no magic smoke. No drama. One minute I power it off, the next, it decides "On" is no longer a thing. It isn't the switch, I checked it.
I expected to find some kind of complicated brushless whiz bang motor controller, but to my surprise, I was greeted by a relatively simple looking single sided board covered in through hole components. Unfortunately the main dip chip does not have any markings. However, I should be able to use the trace map of the board, look for a fault, and then compare it with a working unit after returning it to the big box store and getting a replacement. I get the impression the issue is probably something stupid simple like a blocking diode (or lack there of). I'm just an electronics hobbyist and motor control is outside of anything I'd call a comfort zone. I need and appreciate your thoughts, help, and/or advice. Let's figure out how to fix this thing for fun. I'm sure there are and will be countless people stuck with these units and looking for a solution.

Here are some of the 1 star reviews:
Died Twice
  Purchased at L*we’s. One week later, it didn’t work at all. Swapped it for another. Two months later, this one dies mid board.
If you never power it up, it won't break
  Purchased in February 2020. Did not used it for 4 months. It worked well for 1 project and died while jointing a board on second project. All boards were pine. Since no smoke was released, I will assume, without taking it apart, that the motor control board is no longer functioning
Dead in less than two weeks
  Barely used for woodworking project and died in the middle of a board. Never got overloaded, never got hot, just died! 10-amp motor, my foot!
absolute junk!!
  This thing suddenly quit working half way through the third pass on the first piece of wood that went through it. Just bought it yesterday. Was cutting beautifully until then. Taking it back and don't want a replacement.
good till it quits
I’ve had it for just under 2 months. Worked fine up till it didn't. Took it out today, set the speed and turned on. It worked. Turned off and grabbed a board. Hit switch and nothing happened. Ive never cut more then 1/32 as i feared overheating it, so machine has never been stressed. When it worked it did well.
works fine until is dies.
I had one for a couple months, used it two or three times and it worked well. No complaints. Then when planning some 2x4s it died mid cut. I returned it to L*wes and got a new one. The second one died after 5 minutes of use.

Edit: updated jointer_pcb_map.jpg with a new version, including the correction of a resistor marked 4k7 to 47K, the correction of a capacitor marked 8n2 to "CBB22"/ 100nF, and correctly marking the glass diode as uncertain. 

[floobydust]

The triac tab is discoloured so I think it overheated. You must have good heat transfer to the heatsink or else it will quickly cook and die. I would suspect it might have failed, or maybe the load current-sense resistor 0.1R 7W went open-circuit? What's the triac's part number.
The little glass diode may not be 1N4148, it might be a zener, if you can observe all the numbers on its marking.

I took a look and the mystery IC likely is Atmel U2010B motor control IC.

[asis]

Hi,
In general, the most common failure is the U2010B microcircuit, which provides a soft start to the engine.
These things are installed in many hand tools, in particular in "grinders" and are designed to prevent burnout of the contacts of the start switch.
You solder the DIP16 adapter socket and buy 10 pieces of microcircuits at once, they are quite cheap.
In the meantime, you can use the machine without this scheme.
Or turn on a piece of nichrome in series, wound on some vitrified resistor of high power.
All this while observing safety precautions!
https://datasheetspdf.com/pdf/303345/TEMICSemiconductors/U2010B/1
The resistor values ​​in the diagram may differ from those indicated by you.

[Gyro]

If the U2010B fails repeatedly then it sounds as if it could do with some additional spike protection - a snubber across the motor terminals for perhaps.

[amyk]

Just connect the motor to the mains with a suitably rated switch.

There's a reason (good) motor speed controllers are not cheap...

[floobydust]

We don't know what is failing, the board looks fine just a tad on the cheap side and circuit largely matching the chip datasheet. The best feature is the speed regulation.
The board could be working and it's actually the motor that is conking out, if the brushes are not making contact etc.

I noticed the same speed controller board is used in a bunch of jointers i.e. Delta, Porter Cable. The other boards use two 0.1R 7W resistors stacked, for twice the current settings so that might hint the Sears unit has a tiny motor in comparison.
6" (152mm) Variable Speed Bench Jointer Sears CMEW020 same as: Delta JT160/37-070, Porter Cable PC160JT, motor speed controller board #1345927 (but current-sense resistor is stacked two of 0.1R 7W, not one).

If OP can do this safely, jumper the connections to bypass the board and see if the motor works. Then on a workbench or something connect an incandescent light bulb in place of the motor, and mains to the board to see what it does, if anything. The IC should have around 15VDC across pins 10-11 (or across the 47uF 50V cap) for power. BTA26-600B triac is rated 25A 600V and I find they usually fail short but that would leave the motor running at full speed and everyone says their jointer goes dead.

edit: OP's Sear's controller has the same two-resistor stack as the Delta and Porter Cable boards; I couldn't see that in the first pic.

[jakeisprobably]

Sorry for the delayed reply... busy day, and sorry for the poor lighting and shadows in my garage. This unit has the same double stacked 0.1 R resistors as mentioned, and any discoloration on the thyristor is probably the shadow of my hand holding a phone. - Although the thermal run away issue has a high probability as there was absolutely no thermal grease or pad between the thyristor and heatsink. Nothing obvious is open or shorted from in-circuit testing with a DMM.
   There is a good chance my "1n4148" marking is something else. It would make sense as a zener. I can only see "1N41..." without removing it. The Thyristor is a STmicro BTA26-600 https://www.st.com/resource/en/datasheet/bta24.pdf.
Many thanks for the U2010B hints. I ordered a replacement from within the US and will hit up the grey market for more. I don't think the issue is the thyristor but I have a few ordered as well.
  The motor isn't enormous by any stretch, but the brushes and overall construction look decent. It doesn't look as bad as a typical Harbor Freight Shenzhen Blowout-Clearance-Preorder-Alpha-Proto-Special, but shorted windings and such are always possible with any consumer junk.
  I should have a beginner friendly KiCAD schematic to post soon. I just finished duplicating the Atmel ref. Now I just need to rework it for this board, then I'll start probing around  and testing some more.
-Jake

[floobydust]

Note these phase-controllers use a 0V and -V rail, so circuit common or GND will correctly have (+) polarity on the capacitors. It can get confusing otherwise.

[Nusa]

Note that Sears sold off the Craftsman name in 2017 to Stanley Black and Decker. Sears did retain the right to use the name themselves for 15 years by paying a licensing fee. If you didn't actually buy it at Sears, then Sears is probably not involved. For a planer, odds are good it's basically just another product with minor changes that's been re-branded Craftsman.

[floobydust]

Yes it's not on the Sear Parts Direct website at all, and other bench jointer models show parts "discontinued".
What pisses me off is the speed controller is hard to find and going for USD $175, a total rip off for a wood worker.

Stanley Black & Decker is a mega-conglomerate $31B market cap. Acquired Mac tools, Proto, Bostich, Sidchrome, Beach Toolbox, DeWalt, Craftsman, MTD... Brands Porter Cable, B&D, Craftsman, DeWalt, Irwin, Lenox, Shop Fox etc.  It's all made by one company and then different plastic trim.
"... closed plant and moved all tool manufacturing to Taiwan" kind of thing. I guess I know what brand-names are just shadows of their former selves.

[amyk]

You can always use something like this...

https://www.amazon.com/Newest-Variable-Controller-Electric-Rheostat/dp/B07PYGCHXY/

Not sure if the ratings are anywhere near honest (maybe the original one isn't either) but at least it's cheaper for what is likely very similar circuitry.

[jakeisprobably]

The schematic is attached below. The first version has both the Atmel reference and the jointer motor controller. The second is just the jointer schematic. I double checked it, but let me know if I missed anything. I also updated the trace map in the first post with a few corrections.

Does anyone else think this ground setup is really really bad? The current sense circuit from pin 2 @R3 is sitting right beside the snubber (I think thats the purpose of R9 and C2), and sitting several galactic parsecs away from the #3 connector and the sense resistor termination. Not to mention, 5% resistors for current sensing across a half volt swing, or the 47uF smoothing cap placement... I think they may be using that new AutoCAT PCB designer. Place cat on keyboard and walk away. 

Edit: fixed a designator inconsistency

[jakeisprobably]

I just dug out an old current limited mains tester I built ages ago. I marked "2A limit" on it, but hooked it up to the motor anyways. There wasn't enough current available to run the motor but it turns over a time or few before my tester cuts power. The failure is in the control board.

[jakeisprobably]

I removed and tested the diodes, and both were fine. I also socketed the U2010B in preparation for the replacement.
  I recorded some probing around and the results of directly connecting the motor to mains. Most of the video is probably foolish pointing with pokey things.
https://youtu.be/w7_birbKEOQ
 Indeed, my money is on a bad IC. The real question is: What is killing the IC?

[NiHaoMike]

Add a MOV directly across the motor? That should suppress a big source of voltage spikes. Another MOV right across the switched side of the switch would be a good idea as well.

Or just replace the whole board with a generic motor speed controller.

[floobydust]

The schematic looks OK and the little diode must be 1N4148 because a zener would be 1N47xx or 1N49xx for the bigger 1W ones. It just makes the speed pot less sensitive on the high end.
I don't see a reason for the IC to fail, but no markings makes me wonder if it was from the gutters of guangdong and just cheap junk in the first place. I wonder if the triac still works.
The IC seems to have shorted its power, the 8k2 must still run warm.

[jakeisprobably]

the little diode must be 1N4148 ...

The IC seems to have shorted its power...

Yeah, it is a 1N4148.
  Funny you should mention the shorted power. I tried checking resistance between VS and GND pins earlier when I first removed the chip. Indeed it did show less than 1 ohm. I thought it was unusual to see this with a bipolar chip but I am not experienced enough to say one way or the other on my own. I imagine I should have seen at least one base to emitter drop between the voltage rails.

Edit: Thanks for the suggestions Mike. Anyone know how I should size, or a rule of thumb size, for a MOV in this type of application?

[asis]

Hi,
The most important factor is that in the jointer (as well as in other similar tools) a collector motor is used and the sparking of the brushes provokes large surges of current (voltage) even at the moment when the switch goes OFF, due to the inertia of the rotor rotation and the presence residual magnetization in the rotor and stator of the motor.
And no one thinks that the engine manifold itself is susceptible to both wood dust and natural wear and tear.
Sparking of brushes is inevitable.
The fact that you are doing static measurements, it seems to me, will give little.
This requires an analysis of the internal structure of the IC and an understanding of its work in dynamics.
I am convinced that the whole thing is in the triac driver (16pin U2010B), and its output stage is (probably) an n-p-n transistor.
At least the new U2010B sop16 ICs I have - show an impedance of about 230 Ohm in one direction with respect to GND.
You check pin16 - to call it somehow in relation to GND pin10 and Vss pin11.
The resistance between Vss - pin11 and GND - pin10 on the new IC shows 175 Ohms in one direction.
When you receive new ICs, do the same.
Spark suppression circuit R9 = 100R and C2 = 100nF is ineffective.
R5 = 100R does not protect the open collector from breakdown and affects the entire internal structure of the IC.
The solution to the problem can be the installation of an additional transistor to control the triac Q2, or an optocoupler.
The use of MOV or ferrite beads is unlikely to solve anything.

Of course, I could be wrong.

[jakeisprobably]

This requires an analysis of the internal structure of the IC and an understanding of its work in dynamics.

The resistance between Vss - pin11 and GND - pin10 on the new IC shows 175 Ohms in one direction.

When you receive new ICs, do the same.

The solution to the problem can be the installation of an additional transistor to control the triac Q2, or an optocoupler.

Thanks asis. You shared lots of helpful information.
I have no problem with modifying my device, or getting a different controller. However, my goal here is to show the simplest path someone else can follow and replicate. I have a couple of MOC3041's (opto triac) in my spares. Is this the type of opto setup you're referring to? Something like the first attached image?

I read most of this today:
https://www.nutsvolts.com/magazine/article/triac_principles_and_circuits_part_1
There are several circuits in that article. Maybe at a glance you could tell me if there is a simple topology in that article worth testing out? (fig 18 is attached and the only reference to inductive loads specifically)


On another note, anyone here, feel free to give me a kind nudge when I publish something wrong. Like obviously, the DC power supply is from D1.... I am actually trying to read the datasheet beyond the surface skim needed for a schematic :-)

  I keep having trouble with the "Mains Supply" section. It seems ass backwards using variables that are not defined or explained. Like what is "V_Smax = Maximum Supply Voltage" when there isn't a maximum specification in the "Absolute Maximum Ratings" or in "Electrical Characteristics?" Is it just half the peak rectified line voltage? Then there is the reference to "I _Smax = Maximum current consumption of the IC." I don't see it clearly defined or even vaguely defined.

 Does anyone know what the jellybean alternatives are for a universal motor speed controller with soft start, that has better documentation? I'd just like to save the reference and maybe compare it so I can learn a bit quicker.

Edit: If I understand this correctly (https://youtu.be/Tze1MOZ9Bx4), I can't use the MOC3041 because it is a Zero Crossing opto, and those are for relay-like operation. Something like a MOC3020 is what I need for phase control with an opto...I think

[floobydust]

You've done a great job with the drawings and schematic. Hopefully the original circuit is going to work and the failures are just due to low quality parts. I think it (U2010B) main benefit is speed regulation- something that a vanilla phase-control triac controller does not do. You don't want the machine revving up and then down and varying the amount it's cutting. There are many circuits for constant speed for drills and sewing machines but they are fairly complicated. The soft-start feature isn't really needed.

The original IC design Telefunken/TEMIC datasheet is better than Atmel's but the translation still makes it hard to understand.
The IC's power system is basically a ~15VDC zener diode, as a shunt regulator. You get 1/2 wave rectified mains dropped down by the 8k2 resistor to around -15V.
I don't see voltage spikes causing damage to the IC. A MOV would protect the 600V triac from overvoltage but I can't see the IC benefiting.

The triac needs 50mA of gate current, the IC gives 30usec pulses to turn on the triac. The motor is an inductive load so a 3-quadrant triac (BTA26-600W) would seem better if compatible with the IC but I'm not sure about it.

I did try the cheap eBay "4,000W 220V SCR Motor Speed Controller" boards, $5 and it was such a hassle. The triac died after a month at light loads, then the potentiometer died. I also had to change it from a 240V to 120V design because all the speed control was in the first 1/4 turn of the pot. So I spent $7 to fix the $5 board. But it works fine as a phase-controller and motor speed moves around with load though.

https://www.eevblog.com/forum/projects/makign-a-triac-phase-control-circuit-are-my-values-correct/msg2947016/#msg2947016

[jakeisprobably]

Thanks Floobydust.
 I ordered some BTA26-600BW's...and MOC3020's...and TCA785's - now that's a better datasheet. I have no clue if it could work but the datasheet made me order it.
https://www.mouser.com/ProductDetail/Infineon-Technologies/TCA785/?qs=mzcOS1kGbgcEZNPdCeMI8w%3D%3D

Edit: Not a double/bump worthy post addition, but the idea someone is trying to charge $175 for this board... Well... this should help fix that...
KiCAD .pro files, and a pcbnew board are available here:
https://github.com/Upcycle-Electronics/craftsman_cmew020_jointer_kicad
 This isn't perfect. For instance the mid board component spacing is not a perfect 1:1, they are just close. I did measure and specify the board dimensions, and the location of the potentiometer and triac. These are based on caliper measurements. I also edited the pot's footprint to match the one they used and made a new footprint for the terminals. I did not make a footprint for the trimmer pot, the one I used is close but not exactly the same. I didn't make the double capacitor either. Lastly, this .pro file is based on my notes version of the circuit schematic so it has some additional notation.

[jakeisprobably]

I think the problem with the Jointer motor control design is the Load Current Limitation circuit on Pin 6. With operation mode B/auto-start, if pin 6 is at the nominal 100% voltage threshold at around 6v2, with the main potentiometer at or near the limit on the side closest to Pin 6, this would connect pin 6 across R10/4k7 at around 1.5mA. Pin 4 has a Maximum specified input current of 500μA.
  Also, if R6 is supposed to have a range of +-250mV, I think two 0.1R's in parallel seems way too high for a 10amp tool. If it is really a 10 amp motor, at 250mV it needs a 0.025 ohm shunt right? If I understand the circuit correctly (unlikely;-)
...as configured, the chip should see the R6 voltage hit 250mV at 5 amps during operation. Does this mean it is running into VT100 a lot?

If the issue is too much current from pin 6 to pin 4, I think it should be possible to fix it by replacing the J1 wire jumper with a resistor. This is how it is configured in the reference design anyways. A resistor in this position is intended to set the minimum operating speed.

Edit: At least I think this is a possible culprit. I'm not 100% sure how the 1N4148 diode or the 150n capacitor impact this node when pin 6 is at the 100% voltage threshold. Pin 6 has a negative voltage output. With ground/neutral above this, the diode should not conduct...I think.

Edit: There may be a problem with voltage spikes. I don't speak Russian, and this is a different device with circuit based on the U2010b, but the following is interesting at around 8 min in and is the source of the first attached image:
https://youtu.be/4umgVPMbsBs
A schematic of his circuit is the second attached image. At least I think this is his circuit. The Russian forum webpage does not translate well using FOSS tools and I refuse to use evil-G's spyware browser despite their attempts to leverage it with Translate. I'm not sure what the LM393 in his circuit does.
https://sxem.org/forum/skhemo-temy/693-byt-elektroinstrument-plavnyj-pusk-zashchita-ot-peregruzki-i-obratnaya-svyazyu-po-toku
While I'm adding references, there are notes posted about the U2008b chip at the next link. The chips are from the same family, and his blog notes are somewhat helpful IMO.
https://web.archive.org/web/20200216143903/http://frantisek.rysanek.sweb.cz/U2008B.htm

[jakeisprobably]

So some progress. It's not pretty. I haven't etched boards in awhile and failed to double check my flip sides. I will just add a few brass rivets in the chip pin corners and use a machined pin socket.
 I tried to make it possible to test several configurations using jumpers. I also went crazy with the test point loops just to make it super easy to attach test equipment.  The schematic is even worse than the board, but it's what I have ATM.
  I got a new U2010B in the mail. The resistance measured between -VS and Ground is 680k. Maybe a fake, it's untested so far. I only have one with several more in the mail. It will be a couple of days before I can finish the test board, but let me know if you see anything too sketchy.
-Jake

[jakeisprobably]

Well, I was wrong about over current on pin 4. I couldn't get it to go over 310μA in the same configuration as the original circuit.

 In other news, replacing the U2010b got it running again.

  So...oscilloscope?
Please don't get triggered, or go hunting for the largest font available to write "don't." I've searched. I've read. I probably saw what got written last month...

I have a 300VA isolation transformer half built into a shelved project, and dozens of low voltage isolation transformers, but I think they are too small for this application's current levels.
  I have a hobbyist level 100x probe and I think I could use it without connecting the ground lead to the DUT because the scope is already connected through Protective Earth. I think this means I'm restricted to measuring signals in reference to neutral in this application but I'm not 100% sure I can use PE as one side of my probing circuit. I know the neutral goes to the breaker box, through the breaker and then connects to PE. I know I can't bridge PE and Neutral or I'm effectively bypassing the circuit breaker. So my only option, aside from attempting to buy a diff probe with monopoly money, ...is probe against PE, or float the scope. I want to see the triac switching waveform, how well the snubber works, and the difference between using the three and four quadrant BTA26 600 variants. I plan to do a simple test setup limited to what the scope can auto capture using a stick to hit the capture button and the wall outlet to start and stop the mains AC. The scope is rated Cat I 300V, a Rigol DS1052e. The 100x probe is a Hantek T3100.
  The main reason I'm typing this is to ask if it makes sense to add isolation into the probing setup using a few turns of a couple of wires on a toroid with a terminating resistor on the isolated side with the scope probe?

edit: TL;DR clarity/highlight points

[BrokenYugo]

Substitute the motor for a lighter load, something like a 100 watt light bulb and use your 300VA isolation transformer? A lot quieter that way too.

[floobydust]

In a perfect world you have a beefy enough isolation transformer good enough for the motor under full load.

Check your life insurance policy, floating the scope is very dangerous. If your shit is not together, you will have a bad day. I'll do it with an analog scope (not DSO with SMPS and Y-capacitors) and never to anything lethal for floating potential. Floating it to Neutral would probably start a heated discussion so I will not recommend it, although it can be done.

What I would do:
Leave scope normally plugged in to mains (PE grounded). Probe only using the tip, no GND wire connected, so the scope's reference is PE through its line cord. Without grounding the scope probe, you are limited for looking at low frequency stuff but connecting it (GND) to Neutral is a bad idea because the motor's return current will rip through the scope to PE, and damage it internally.

Ch. 1 on point of interest (should range from mV (shunt) to mains V plus transients- using x100 probe)
Ch. 2 on Neutral (should be a few V under motor load)
Set scope to math, subtract Neutral from point of interest.
AC Line trigger the scope

The motor is a large source of EMI due to the brushes arcing. I don't get spectacular results with Rigols in high-EMI lab setups. Neutral will also be noisy and a few volts different than PE. If your Neutral is quiet enough, then you can just ignore it and free up the second channel.
The IC's circuit is awkward with referencing everything to -VDC as the common, so drink lots of coffee. I wonder if the IC really does decent load/speed compensation.

Your tests can be very interesting, but just be prepared to abandon ship if the test equipment isn't enough, rather than take on high risk. It's not worth it.

[jakeisprobably]

Well I have lots to share but I need to prepare and polish the visuals first. I'll try to avoid adding controversy about how I probed things. Unfortunately my (unhacked) DS1052E is not fast enough to see any transient voltage spike beyond the effects of the default snubber values. I didn't test it with the snubber disconnected entirely though. With it connected I didn't see anything major.
 Both the BTA26-600B and BTA26-600BW work, the 'BW that I tested showed a slightly higher Vpp for whatever reason. I didn't investigate further, but I have the waveforms saved.
  I have a spreadsheet with resistances measured to and from GND and -VS to all pins for all chips so far. I now have three tested and verified working chips from 2 sources and three defective chips (not including the Jointer's) with more still in transit. Indeed it seems there is an issue with IC quality if any conclusion can be made from such a small sample size. Three out of the five I received two days ago each had a different failure mode. One has shorted power, one has no feedback connections on pins 4..7,  and the last has low resistance values on the compensation amplifier/buffer IIRC.
  Interestingly, the set of 5 chips have sharp-ish top epoxy, the kind I usually associate with refurbished chips, but the casting stamp designator format and dimensions on the bottom of all five are matching just with different letter/number combinations. I belive they were all from the same manufacturer. They don't match the Jointer's chip cast markings, and nothing matches the third chip I have either. These faulty chips are exactly that. They all have pins with expected values on various pins with a failed block somewhere on the IC. Overall, I'm building up a map of what the "good" range of resistance values are and measuring everything as soon as I receive them. I hope I see some pattern in any future failures if I have any.
 I will have more visuals to come, but it looks like my conclusion will be to just add a socket and swap the chip.

[wizard69]

Honestly I'd chalk this up as a lesson learned.   These days everything with the Craftsman logo on it is junk and that is being kind.   That is really sad to say but I wouldn't trust anything marketed as Craftsman.

In any event you probably want to get you jointer to work, the first thing you need to determine is what sort of motor is in the machine.   Figure that out and then go out and buy a real motor controller form the likes of KB controls or similar.   Do that only if you think the motor will last which it likely won't.

Your best bet would be to simply buy a single phase motor and run it directly off the line.   Even here you may need to change the belt drive and the guarding around that.

Now given all of the suggestions above to get it running, ignore those and return the piece of junk and demand your money back.   There really is no excuse for such crap being shoved down our throats.   When it comes to tools sometimes you will not have a local option, you will likely have to order a decent machine via other avenues.

[jakeisprobably]

The machine is made by Shinmax Industries https://www.shinmaxindustry.com/
All are the same:
Delta 37-070
Porter Cable PC180JT
Craftsman CMEW020
The first part of the above part numbers is the sticker company, not to be confused with actual sticker manufacturing companies. These are sticker sticker companies. Note: sticker sticking is also subcontracted. As far as I know ""Craftsman"" has always been a sticker sticker subcontracting kind of brand. Like most of the 80's-00's stuff was made by Ryobi (aka all Craftsman part numbers that start with 315.xxxx. IIRC the 113.xxxx stuff is most of what was "Made in the USA" and was the part number prefix for Emerson.
  Venture capital is what it is. They are counting on consumer ignorance and resistance to change in order to market their overpriced junk. If that bugs you more than it does me, maybe you're motivated enough to beat them at their own game and seek out the best value products for you, forging your own path. For instance, when Sears sold Craftsman stuff made by Ryobi. Ryobi made better stuff at a better price under their own brand.
History repeats. Shinmax's house brand is Cutech: https://cutech.tools/
Their 8 inch wide bed, spiral cutter, and bed extensions model is listed at a really nice price, and is very tempting if shipping is reasonable IMO.

Attached is my spreadsheet with test results for 17 total U2010B chips tested in and out of circuit. Maybe someone that knows more about silicon can comment about the range of values and anomalies. Is the range of resistance values seen here normal?
   I am really surprised the chips labeled ALI-12 and ALI-14 worked. Chip 12 appears to have major issues with the sense pins and 14 has no connection on Csoft at all. Of the bad chips on sheet 2, my favorite is the one with no connection to the ground pin.
The spreadsheet was made with Libre Office, but also saved as an xlsx file for 'doz users, and/or there is a PDF.

[Ground_Loop]

I understand you're trying to fix this for fun,  but please understand that there is absolutely no reason to have speed control on a jointer.   If you really want to use this tool then the motor needs to be directly connected to mains.

It's a criminal level shame what accountants did to the Craftsman brand.