A reasonable expectation when buying a new movement from a watch materials house is that it will be fit for purpose. Last year I found myself on a slightly feverish run of mix ‘n’ match self-build projects constructed around Seiko Instruments NH35/36 and NH34 movements.
The NH34 is essentially the same as the three-hander day/date NH36 but foregoes the day complication in favour of a 24-hour wheel. Â
Three of these diversionary projects featured the NH34, and all came together relatively easily with the only off-piste element being the requirement to modify the dials to accommodate the larger diameter tube of the 24-hour wheel.
However, one of the three completed watches presented symptoms, the cause of which I struggled initially to diagnose. In most respects, it functioned as one might hope but the movement showed a distinct reluctance to giddy-up from a standing start. A sure sign of a happy movement is that it will run from its unwound state without a great deal of encouragement: sometimes just picking up the watch will be enough to get it ticking away, but this specimen refused to start without at least a few hand winds of power into the mainspring.  Some back-and-forth trouble-shooting revealed no obvious clearance problems with the dial or hands and so I turned my attention to the movement.
A timegrapher measurement revealed underwhelming amplitude on a full wind and some very poor and inconsistent figures on less than a full wind.Â
The headline amplitude of just shy of 250 degrees might not seem like a completely terrible result but the other movements were achieving 270+ degrees and weren’t displaying the same reluctance to run on low power. The unavoidable conclusion at this point was to suspect that the movement may not be in a state of health that one might hope for from something purporting to be box fresh. Some exploratory investigation appeared warranted. From the train side and with the winding weight removed, the movement appears indistinguishable from its three-hander brethren.
Perhaps the most obvious potential impediment to amplitude and free running of the movement is the escapement and that is where we visit first with the balance removed.
No smoking guns just yet but on removing the pallet bridge I was surprised to see that the upper jewel hole had been lubricated.
It is common practice to leave these holes unlubricated to minimise drag on the escapement and to limit the scope for dirt to congregate at the (beating) heart of the movement. However, counter to this received wisdom, the technical guide recommends lubricating both sides with a normal quantity of Moebius 9010.
Perhaps I am revisiting my thought processes at the time with the benefit of hindsight, but the potential folly of this approach was confirmed once I’d prised the reluctant pallet fork from its lower bearing.
How about that for a smoking gun? What then might be the cause of the accumulation of brown gloop? A clue comes from a close-up of the lower pivot of the (now cleaned) pallet.
Is this the cause or the effect? Could a lightly corroded pivot have added material to the oil which then turned into a contaminated slurry? Or might the lubricant have become tainted with another contaminant which then eroded the pallet fork pivot? Either way, it is clear that we have identified the source of the problem. The only way to rectify the issue is to dismantle the movement, clean it and replace the pallet fork. I don’t propose to document that process in detail but it’s worth pausing once or twice to highlight noteworthy observations. The first of these is to note a signficant design quirk of this family of movements which is that the setting parts are located on the balance side of the movement.
With everything out of the way, we can get a better view of the contaminated pallet hole jewel.
Given that the primary reason for the poor performance clearly lies in this area and that the movement is otherwise new and unused, I didn’t feel the need for a belt and braces service and so some of the peripheral parts were spared the ordeal of passage through the cleaning machine.
The reassembly process is documented in part below, top left to bottom left in a clockwise direction.
The conundrum I reached at this point was to identify the most economically attractive route to the sourcing of a replacement pallet fork. A new replacement would have set me back more than £20 including VAT and delivery (which is frankly outrageous given the cost of a complete movement) but for the same amount I could buy a complete NH36A from AliExpress (including shipping from China and VAT) and harvest the part from that. So that is what I did.
Out with the old (below left) and in with the new (below right).
Unable to break my force of habit, in fitting the new pallet fork, I resisted lubricating the upper and lower jewel holes. Here is it, back in action, in situ.
The proof of the pudding is in the eating.
This somewhat monstrous amplitude settled down to high 280’s dial down but importantly, it now starts up with the slightest shake from slumber and I am content to slip it into circulation from time to time without that nagging feeling that there is something that I should be attending to.
