As a start, you may want to read my quick introduction here
I did a full strip-down of this movement and properly cleaned its parts with lighter fuel. Before the "service" (if you can call my first real service of this movement like that) the amplitude was at 160-170°, which wasn't exactly great and the watch gained around 60-70 seconds a day (according to the time grapher).
I tried to follow all the good advice from the watch repair lessons, wore e.g. finger cuts which was quite difficult in some situations. In general, I'm quite happy with the result. The movement looks clean and everything was smooth (e.g. train of wheels) when I assembled it. However, the result is not yet good on the time grapher.
Coming to my issue with this movement, it looks weird on the time grapher where the amplitude travels between 190° and 240° and without touching the watch the watch moves between losing/gaining -35 and +40 seconds/day "in waves" on the time grapher.
Full disclosure on the issues that I had (and introduced myself):
I scratched the hour hand, I know that this is not acceptable, but nothing I can do now apart from trying to polish it when I reassmble it (it is my watch anyway and I don't wear it, but still not acceptable)
I damaged the upper pivot of the pallet-fork during re-assembly and therefore acquired a scrap ETA 2451 (Dugena) movement from eBay as an attempt to find a replacement pallet-fork. This wasn't successful, as this damaged movement also included a damaged pallet-fork thus I bought a new pallet-fork for ETA 2451 via eBay (I currently don't have any way to order at Flume, as they only sell to corporates/business, not private persons)
I wound the mainspring in by hand and honestly this wasn't something that went well. I've already tried numerous times to get hold of a mainspring winder, but they're quite sought-after on eBay and up their price in the last seconds of the auction by exponential means
The issue I had with the mainspring was, that I didn't do this before and I first tried to wind it by hand and hold it by hand and later on put it into the barrel fully wound. This wasn't only very exhausting, but the skin of my fingers (yes, I couldn't use finger cots at this attempt - as it drove me nuts) at thumb and index finger started to be injured (at least you see, there is a lot of passion :))
The attempt afterwards was more successful where I cleaned the mainspring again and wore finger cots and wound it into the barrel directly starting at the outside walls
Note that the mainspring didn't look very good, when it lay on my workbench (it looked "too set") which was likely caused by my first winding attempt
The balance pivots (incabloc endstones) are not lubricated yet
In general the lubrication looks much more simple job when I watch the videos by Mark (I take it as a positive, that the amplitude reading is better after my service)
I used Molykote DX on the clutch of the keyless works
I used Moebius D5 on the heavier friction points
I used Moebius 9010 on the more delicate friction points
I haven't lubricated the pallet-fork pivots
I didn't (yet?) lubricate the entry & exit stone with Moebius 9415 (ETA 2451 is only a 18000 bph movement)
I didn't lubricate the balance / incabloc settings yet (this is what I'll do next)
Now that you know the issues that I'm aware of, the following is my list of things that could cause the issue I see on the time grapher. I'd like to hear your opinion/advice what you believe could cause this weird reading.
Magnetism, I should demagnitise the watch movement (which I currently can't, as I don't have a demagnitiser yet - am about to acquire one via Amazon)
I might have damaged a wheel during cleaning (all looked good to me) and it doesn't run evenly good, depending on the position in the train of wheels
The mainspring is causing this issue, as it doesn't provide "steady power" and thus causing this issue (I'd more expect an issue with the power reserve though)
The beat error is too high for my taste, but I have difficulties to adjust it (I'll look at this in detail again, it is not the source of the timegrapher waves in my opinion)
Some words on the movement:
The Doxa 115 is a 11 1/2"' movement, soley based on the ETA 2451 (I didn't find any difference, when I looked at the scrap Dugena ETA 2541 movement)
The movement runs at 18000 bph and the automatic appears to wind it in both directions (as far as I can tell)
The mainspring is this one ZF-869 which I'll try to get hold of to replace it
I guess that this movement has been discontinued as you need to fully disassemble it to replace the mainspring, so it is not as service-friendly as later more modern movements (However, I do this for the fun of disassembling it, therefore I don't care that much :))
I'll update here when I've lubricated the incabloc setting endstones, but I don't expect to see an improvement. (Hopefully, I'll be proven wrong by myself ...)
I also noticed that the movement might be under tension due to the winder/stem being pressed into it by the time grapher, will check if this makes a difference.
Have a good weekend! I hope you liked this post