Why I keep cross-referencing CNFans spreadsheets
So here's the thing: the CNFans spreadsheets are a rabbit hole. On one tab you see folks hyping the PT5000, on another, the Seiko NH35 is king. I've spent late nights comparing notes, actual rate results, and service chatter. And because I hate guessing, I pulled in COSC specs, independent timing logs, and my own timing runs with a cheap timegrapher.
I personally think the real value hides in the boring details: beat rate stability, service intervals, and how a movement behaves after a year on-wrist. Let me break down what I've found from three popular spreadsheet sources that most of us bookmark.
Accuracy: what the data actually shows
One spreadsheet leans hard on the Sellita-clone ST2130 (28,800 bph) as "best accuracy per dollar." Another pushes the PT5000 with a similar beat rate. The Seiko NH35 (21,600 bph) shows up on nearly every budget list. Let's look at actual numbers:
- PT5000: Rated at -10/+15 sec/day by the maker. I've seen at least 3 posts on Reddit showing +3 to +6 sec/day after basic regulation. A 2023 WatchTime lab test on a microbrand using PT5000 recorded an average of +5.4 sec/day.
- ST2130 (Hangzhou 6300 series): On paper, -5/+10 sec/day. My own timegrapher readout on a brand-new piece was +4 sec/day dial up, +6 crown up. There's a long-term log on TZ-UK showing an ST2130 holding +7 sec/day after six months.
- Seiko NH35: The spreadsheets call it "workhorse" but few expect COSC numbers. Factory spec is -20/+40 sec/day. Most arrive around +12 to +18 sec/day unregulated. After a quick regulation, mine sits at +8 sec/day. No, it's not Swiss, but it's predictable.
- Miyota 9015 (also listed in some sheets): Spec -10/+30 sec/day. A Fratello review sample ran at +7 sec/day in multiple positions.
- NH35: 24 jewels, Diashock, and generous oiling points. Seiko publishes service docs, and parts are everywhere. I've had one survive a minor drop with no rate shift.
- PT5000: 25 jewels, Incabloc-style shock. Early batches had noisy rotors; later runs seem quieter. A watchmaker in Hong Kong reported fewer amplitude drops after 12 months in newer stock.
- ST2130: 28,800 bph clone of ETA 2824-2 architecture. Parts interchange with some ETA bits. Anecdotally, I've seen two reports of hand-wind gear wear after a year, likely due to dry lube from factory.
- Miyota 9015: 24 jewels. Thin profile but known rotor wobble. That wobble is annoying, not a death sentence.
- NH35 builds: $45-$80 for the movement inside a $120-$180 watch. Lowest cost of ownership and easiest service path.
- PT5000 builds: $70-$110 movement inside $180-$260 watches. Better accuracy potential, moderate service access.
- ST2130 builds: $80-$120 movement inside $220-$300 watches. High beat smoothness, but lube consistency is brand-dependent.
- Miyota 9015 builds: $100-$130 movement inside $240-$320 watches. Thin profile, decent accuracy, louder rotor.
- Daily beater: NH35. It'll run a bit fast, but it'll run. Cheap to fix.
- Accuracy chaser under $250: PT5000, regulated. Find a seller who shares timegrapher shots.
- Smooth sweep fan: ST2130, but ask about service history or buy from a microbrand that does QC (some list their testing data).
- Thin dress watch: Miyota 9015. Ignore the rotor buzz, enjoy the profile.
Now, COSC standards for chronometer certification are -4/+6 sec/day. None of these budget-friendly movements are COSC certified, but both PT5000 and ST2130 clones can be regulated to live inside that window. The NH35 usually cannot without luck, yet it's consistent enough that you won't see wild swings day to day.
Reliability: parts, lubrication, and shock tolerance
Reliability is where spreadsheets differ. One community sheet warns about early PT5000 rotor issues from 2018. Another praises new batches post-2021. I dug into service threads and Caliber Corner notes:
Honestly, the NH35 wins on sheer parts availability and forgiving tolerances. PT5000 and ST2130 beat it on beat rate and smooth sweep. The thing is, if you don't have a local watchmaker comfortable with Chinese calibers, the NH35 becomes the safer bet.
Longevity: who keeps ticking past five years?
I wanted to see real timelines, not just spec sheets. So I skimmed YouTube teardown videos and forum logs where owners reported rates at 12, 24, and 36 months.
For the NH35, there's a three-year ownership thread on Watchuseek where most users kept +10 to +15 sec/day without service. That's boringly solid. A microbrand using PT5000 published their own internal warranty returns: less than 2% over two years, mostly crown tube issues, not the movement. ST2130 clones are a bit mixed; some early failures tie back to inconsistent lubrication. Miyota 9015? Plenty of five-year-old pieces still within +10 sec/day, though the rotor noise worsens with age.
And yes, COSC-certified ETA 2824-2 beats them all in stability, but it's rarely under $200 in CNFans lists. At the end of the day, longevity without service also depends on how you wear it. If you rotate watches weekly, even the fussier PT5000 will likely be fine for years.
Value propositions from the spreadsheets
Let's map the movement picks to actual value, using current street prices from two popular CNFans sheets (as of this month):
The kicker: If you factor in a $120 service every 4-5 years, the NH35 looks even better. But if you care about second-hand sweep and tighter daily deviation, PT5000 and ST2130 feel worth the extra $60-$100 up front. I lean PT5000 because new production seems tighter; I personally think the rotor improvements fixed the old complaints.
My short list based on use case
Look, no two wrists are the same. Here's how I'd pick from the CNFans spreadsheets if a friend asked me over coffee:
Now, this is where it gets interesting: some spreadsheet curators started adding QC columns—amplitude, beat error, positional variance. Those sheets, in my opinion, instantly boost trust. If a seller posts +4 sec/day across three positions, I'm way more likely to pay a small premium.
Long story short, the CNFans spreadsheets give you a head start, but pairing that with published specs (COSC standards, manufacturer tolerances) and real timing logs turns guesswork into a plan. I won't tell you there's one perfect pick. But if you match your tolerance for rate drift with your willingness to service, the value choice becomes obvious. For me, that’s a regulated PT5000 for anything sporty, and an NH35 when I want set-and-forget.