time for a bit of update, today I did my next step on the modified Dahlander, with some plusses and minusses, but let's take it step by step.
First an answer to Weber, regarding the connection of the wiring. I measured the coils after I broke the internal star point -> the connection is like the second picture in your message, with the outer connections crossing, and the middle one just passing straight.
Step 1: I opened the motor again, and tried to open the internal star point. Is seemed to be soldered, but even with a gas burner, no sign of melting, so I decided to cut it off. I hope I am not going to regret this step.
Step 2: connecting new leads to the opened windings. I checked with a DMM, and indeed the wiring was as we thought. It is pretty tough getting a proper reading on the DMM though, and I still have the feeling not all resistance values were the same.
Step 3: connecting the leads together (switching two original leads to come to the standard mounting order instead of the crossed over Dahlander order). Furthermore building three short bus bars to get to delta mode. We used flattenend copper pipe for this, works quite well for our current purpose.
Step 4: connecting to the VFD, and starting up -> it works !!
Step 5: but .... it makes much more noise, and squeeching tones. Even up to 50 Hz is is still screaming a bit. I will upload a movie to youtube, so you can judge for yourself.
Step 6: so, listening to your advices, I decided to start an AMA. However, this has not been succesful. During the Sensorless vector control phase of AMA, somehow the current became such that the house fuses did not cope anymore.
Step 7: tried to do SLV with some approximate motor parameters, however, the motor then makes a big jump and goes into error mode, so not succesful
Step 8: again doing AMA, now it got further, and also in the movement area some progress, instead of just vibrations, the motor walked about one meter
, very scary.
So after this, and some more blown fuses, I have decided to call it a day.
The issue I have can have many reasons as I see it now:
- some connections that I have made are not correct, so the current is not equal
- the windings are not meant to be used like I use them (for your information, I have 36 slots for the stator windings, if that is helpful for someone)
- The starting values of AMA (base freq, power level, motor voltage, ..) are not good to come to a proper autotuning
Also I noticed that the motor was a bit warm (hand warm), which is strange, considering that no power was demanded from the motor. Or is that normal during an AMA session?
I have thought that I should make one step backwards, and go to the original wiring again (very simple now with the external leads to the star point). THen do an AMA again and see how that works.
If that is going well, then at least it seems that my new connections are sort of OK (they are temporary, but should still be proper for safety and performance testing).
From there I was thinking that I take the parameters from that AMA,and do the conversion from parallel star to parallel delta numerically, entering these values by hand in the VFD. And the do a run-up again (and pray
On a lighter note, I noticed that my VFD is not a SJ300-550 (55 kW nominal) but a SJ300-750 (75 kW nominal)!! THat is somehow a miracle, as I paid for a 550, but I will not complain I think.
I will put some pictures here tomorrow, when I have found the cable of the camera.