Vintage Juki industrial sewing machine DDL-555-5 repair

[max.wwwang]

Got my dirty hands on a lovely Juki industrial sewing machine. This is a perfect toy for me as someone who likes both mechanics and electronics. Its head unit is in reasonable condition with signs of wear and tear (it's not a casual machine so this is not a surprise). But the rest, such as motor, control box (the box below the motor under the table), CPU mox (the robot like small box above the table), are all in mint condition.

This was after I repaired, or tuned, a handful of domestic sewing machines and looked into another Brother industrial machine. This one impressed me and I really want to get it running the way it should. Its look is sleek without exaggerated masculinity (though it is powerful). Its electronics design is elegant and compact. Its design is highly modular, with plenty room of for different sorts of adjustments, and other beautiful features. The only aspect that loses to the Brother is the pedal sensor module. Well I've got into too much detail that you might not be interested.

When I got it, the motor ran when turned on, and the head unit also ran for a second or so without pressing the pedal then stopped. After that no reaction to the pedal. After some fiddling around, I don't even exactly know why it started getting 'better'. The motor always runs well when turned on. The needle responds to the pedal, running when pressing the pedal forward, and accelerating when pressing hard. The thread trimming mechanism (driven by a solenoind) works as well but the timing is incorrect so breaking the needle (the knives action when the needle is down, which is wrong). Although it would be great if this can be fixed it is not the biggest problem, because this feature is not essential I can simply disable it if cannot be fixed. The biggest problem now is that the needle does not stop immediately when the pedal returns to the neutral position. It stops, but not immediately. Without this being fixed, the machine is useless, because it does not stop sewing when you want it to stop.

I've tried all I can, including disassembling and reassembling the pedal sensor module, the synchroniser module, tuning the delays with the pods inside the control box. I've also read the service manual carefully for many times and have exhausted what I can think of. In case someone has experience or is interested in helping me crack this. It's quite a challenge, so this is an invite!

The service manual is available here https://s3.amazonaws.com/a.teamworksales.com/JUKI+ENGINEERS/DDL-555-5_100%26200%2CEngineer%27s+manual%2C(No.I-14).pdf. The level of detail and clarity is impressive given the age of the machine (I think its manufacturing was from the end of the 1960s to probably the 1980s).

Since it's very rare when a sewer is interested in electronics, there is little, if at all, information on youtube related to this.

P.S. Well I said I've tried all I could, but I've not really got into the electronics yet, i.e. tracking the signals in relation to the pedal positions etc etc. This is probably what I should do next. It's only not as wieldy as a little box that I can put on the desktop and toss around.

[max.wwwang]

Industrial sewing machines of this age usually come with a clutch motor. This is no exception. However, this is a more advanced electronically controlled clutch motor, like my Brother. When turned on, even when the needle is idle, the motor runs constantly. There is a clutch between the flywheel (always running) and the pully. There is also an electronically controlled brake along with the clutch. The speed control of the output side, the pully, replies on the delicate control of the clutch and the brake, and probably the speed sensing with the synchroniser on the head unit. Here is the relevant part of the service manual:

[max.wwwang]

The pedal sensor is a potentiometer, which is a design far less durable and far less elegant comparing to the same module of the Brother. Just to sidetrack a little bit, here is the module of the Brother. It will last virtually forever unless the electronics fails because of no possibility of wear and tear due to its contactless design (detecting pedal movement with hall sensors - one is DN839).

[max.wwwang]

It was slow. Because the machine is bulky I had to take the control box off to have a close look and to test I had to put it back under the table.

But some progress was made, in both directions. That means something went worse but overall the path is clearer and a replacement component has been ordered.

First, isolation - or narrowing down - the problem or where is problem may lie. This machine has many great features. One is that it can be configured without the CPU module (the robot like thingie above the table) and the machine can function almost the same (only without those additional features provided by the CPU module) with only the control box hanging under the motor. This is configured, the problem persists. This means the fault is in the control box, probably not in the CPU box. Also traced down where the voltages come from driving the clutch and brake solenoids in the motor. This is what gets clearer. What gets worse is that, after connecting/disconnecting a couple of times (as well as a bit of disassembly of the boards, cleaning on the way, etc.) the problem becomes - whenever the machine is turned on, the head unit runs non-stop and won't stop. This means not only the voltage for the clutch was not cut off when the pedal returns to neutral, it now provides the voltage all the time, so the pedal does not function at all now. Measuring from the output port of the control box gets a 34V DC or so output, which comes from a 2SD1087 Darlington power transistor. Measurements between its pins are suspicious and are very different from another identical component (for the brake solenoid). This is not believed to be the culprit (and hopefully the only one). I hope the previous failing-to-stop problem was only due to the problem of this chip, not reacting to the signal given to its base the way it should have. Now I'm waiting for the replacement part with fingers crossed. Alas, this will take at least one month for Aliexpress's delivery to come!

I can't help saying this again, this machine was beautifully made!

[max.wwwang]

Another thing I want help on is - a NEC D1514C is apparently the brain of the control box but, strangely, its datasheet is not readily available online. Does anyone know something about this MCU IC?