In today’s hectic and superficial world overcrowded with bogus experts this channel (you) is a real gem. Old-school way ; proper, thorough, real-world explanation for both beginners and wannabe masters in the art of watchmaking. No BS, right to the point, your presentation is involving and compelling. I went through all of your lessons and I have no doubt that I will return to them many times to come. My deep bow for will and effort to spread know-how. Please, continue as long as you can. I trust that you have much more to share.
Wow brother thanks so much. This whole thing is keeping me pretty busy these days and I appreciate the heart felt words from you. I got tons more coming. See you again.
Just a note of thanks for your well-structured and very clear videos. Like many I watch lots of U-tube material, and you are in a very small group of the very best, well qualified, patient, very generous and clear content posters. I, and I'm sure thousands of others, have gained hugely from your work and I'm massively grateful, thanks!
Yes, agreed. Thank you!!! As an amature, trying to learn as much as I can as a hobby, who knows what this could turn into. Thank you for your insights.
I have the exact same model and could not live without it, a Watch on the wrist can also tell a different story. A timegrapher is a good indicator of how a movement is running for sure, they can also give a slightly different reading with an uncased movement. Great video thank you.
I often find my watches on the wrist are performing much better than on the time grapher. On the other hand, there's a good deal of variation between the different positions that presumably compensate for each other. I've also got that model.
You're exactly right. For hobbyist watchmakers who gain satisfaction from making an Ebay vintage come back to life and run well again, the Weishis are absolutely fine for gauging skill level and accomplishment. It's doubtful it ever happens, but if this hobby ever graduates to working on more expensive watches, I'll upgrade to a better timegrapher. In the meantime, my old Gruens and myriad ebauche watches will do just fine with a Weishi. Thanks, Alex. Your videos are always thoughtful and well presented. I've learned tons from them.
I’ve found the Weishi timegrapher to be excellent for detecting strange noises that occur when something in the escapement is touching something that it shouldn’t. When the tracings are irregular or wild, it’s time for the microscope even if it’s keeping time in that position. It’s been a great tool to have! Another great video, Alex!
It's amazing that these tools exist for the modern watchmaker. I recently re-read "Longitutde" by Dava Sobel on the history of John Harrison and his quest for an accurate timekeeper to calculate Longitude for ships at sea. Mr. Harrison, I'm sure, would be amazed at the current state of mechanical movements and the tools available. Great video as usual. You are as great of an instructor as you are a watch maker. Thanks!
John, I am sure he would have been amazed but what about the guys from the early days of the Railroad watches that had to regulate watches in 5 positions. They would have loved it.
Thanks for the video so quickly Alex. 2spd is nothing to worry about for most watchmakers. Finding the lift angle on a clean movement and calibrating the 1900 with 180° on the balance was fun.
Great vid! This is almost identical to those of us in serious home theater calibration. The supplied mics with low and mid tier receivers are good enough to get things very close, but for highly accurate top level results you need the calibrated mics and the software that can use the calibration files.
First: Thank you so much for your videos, I learned a lot with them and just started taking apart my first movements :-). Just a little comment on this one, as I think you make quite an error with the comparison here: You compare a one minute mean value of the escapement analyzer with a momentary value of the Wheishi (which is what is displayed big) so u compare two different things, especially as the value always has some scatter in it. If you take for example the mean of the last 4 12-second means of the Wheishi (+6 +6 +3 +3) you end up with a mean of 4,5 s/day, which is quite close to the mean of the escapement analyzer. If you want to compare rates I think it´s good for both systems to have them sit for like 30 seconds and then compare the (mean) readings for the same time period. For the Weishi the default value for the means is 12 seconds, so take 5 of the smaller shown readings and calc. the mean. Same is valid for amplitude, but there the Weishi only shows the momentary value, so it will be hard to compare
You are welcome. I appreciate your input but let me make a couple points. On timing machines, you don't consider the numbers from the "settling in" period. This would make your data wildly inaccurate and useless. Mean values are pulled from the last rate of the timing period. Usually only the highest of quality movements can maintain a stable rate or amplitude for any length of time. Most will move up or down for various reasons related to the power transmission and is normal. WIldly fluctuating numbers are a signal of a fault and need correction before regulation. There were two points to this video. The first is that these machines are close enough for a home watchmaker needs and the second point is that you don't want to guess as to what the lift angle is. -Alex
The Weishi is a great hobby tool for the money. If you are attracted and inclined to the absorbing and myopic universe of watch tinkering being able to see your cleaning make an immediate difference is very satisfying. Great channel BTW.
You're welcome! For repair any app or timing software that doesn’t measure the amplitude is pretty much useless. It’s would be like trying to diagnose a quartz watch problem without testing the battery.
I have a Weishi 1000, I've done my best to test it by just obverving how accurately my watches (Planet Ocean 600m and Superocean 44) run over a few days compared to what the weishi is telling me and also by comparing its results with the watch accuracy meter phone app, everything seems to agree pretty well. I'm not a watchmaker or horologist I wouldn't open my Swiss watches and try to regulate or service them but I just find the Weishi handy to track how my watches are running and whether they need a service or not, I also have a few friends with watches who ask me to test theirs occasionally for the same reason. From the information I've read it seems the newer Weishis are a lot more accurate than older models they have improved the electronics and firmware, for instance an older weishi 1000 couldn't detect the amplitude of an Omega Coaxial movement properly but my newish 1000 seems to work properly in that regard.
remember too that there is software in both of those devices that will make them not completely comparable. I was a research scientist for years and whenever we compared data from different devices, we would frequently have to utilize another program to directly compare the results. For example, both of your datasets have 'smoothing' applied. That, in itself is an equation. What would be better is to open the movement, and utilize optical references to determine BA and AMP, and then run a multi-day test to judge accuracy, then compare back to both programs. I would wager that you'd then have 3 disagreements of data.
Well, pretty much everything you just said his way over my head so let me just say this. This is a direct comparison of the ability of two separate sets of software, measuring the same thing. It’s barely to show the accuracy between one set of software that is highly calibrated versus a common Chinese variety timing machine.
Thanks again Alex, as always with measurement tools they need to be calibrated regularly but I don’t believe there is a facility available for the weishi timegraphers. I have experienced very expensive and regular requirements for calibration from my days in electronics many years ago. I don’t believe that with the tolerances needed in watchmaking it is too big a deal, the original railroad watches there were no timegraphers available to them.
Have to add, I also worked in electronics companies where calibration was less important and our repairs were fit for purpose. I think us amateurs to watchmaking can easily get too caught up with seeking perfection with a timegrapher rather than accepting that we have got it into specifications that a watch was originally designed for.
Totally agree Tom. There is a calibration port on the back but I don’t think I’ve ever heard of anyone calibrating one. I’m pretty sure that are primary used in the factory.
Thanks for another great video. In my experience the Weishi should generally be taken with a grain of salt, and perhaps even the expensive Swiss timing machines. Also in my experience the gain setting affects the amplitude reading quite a bit. To be sure I record a slo-mo video of the balance. Anyway, the Weishi is good enough to make sure things aren’t completely off.
Great point Its just a tool. Accuracy on the wrist is the bottom line. This was really for the guys who say just use 52 as a default LA. Its close enough. Well not when troubleshooting.
That's a good and honest comparison Alex. I recently bought the Weishi 1900 and it's a great little machine and is a confidence boosting guide that you are doing something right when comparing a pre and post service. Just finished servicing an ETA 2783 that indicated low amplitude on the Weishi. Post service it's running like a champ with good amplitude, beat error and rate. One question/observation - I do notice a difference in readings depending on whether the watch is cased compared to uncased or in a movement holder as compared to sitting directly on the stand. Do you think the microphone is affected by resonance from, for instance, a movement holder?
Great video Alex! Curious how the Witschi machine would stack up against the Weishi -- basically if they're truly identical internally, or if the components in the Swiss version are better, or perhaps calibrated to a higher standard? That software you're using looks cool too, but how do you hook up the microphone?
The Swiss machine is going to be manufactured to higher standards being that’s it’s used professionally and it can also be calibrated. The Weishi has a port for calibration but I don’t personally know if any company calibrates them. The I have a mic stand that plugs into the computer.
The Watch Escapement Analyzer and Timer version 2.172 looks interesting. Is the entry level $400 system what you use or do you use a different variant?
It’s just as important to know precision (test/retest variability) in addition to accuracy. If precision is good the inaccuracy can be adjusted and the measurement will be reliable.
There was a you tuber who stopped posting once he went pro and he said when he got Omega certified , Omega told him to replace his weishi with a witchie 😮
Very good video as always Alex but I can't help thinking "how do you KNOW the expensive software is more accurate than the Weishi? I have a very low budget solution .. a small TBGC device from Ebay (about £30) communicated to by two open source software graphers both give similar but different readings to each other. Again, are any correct? For what my opinion is worth I think they all give some indication of a running watch and are useful for seeing potential improvement post servicing
Sure they do. I am not saying they are bad by any means. The main take away might be to make sure the lift angle is correct. Also my timer is calibrated and shows the deviation from the Atomic clock. It is calibrated to +/- .09 spd. You can see it in the bottom right of the screen
Thank you Alex. I have been looking at ebay avidly for a timegrapher. I think a 1000 would suit me, however would you consider it prudent to go for the 1900. Cost in the UK are £140 for the 1000 and £170 for the 1900. Makes a month extra saving for me. By the way, I have completed building my first two watches. Happy bunny.
Excellent video - thanks. The very short test duration will also amplify any differences shown by the measurements. Can your calibrated software record and average readings over a longer time period as well as provide instantaneous readings? It would be interesting to see the test repeated over a 24 hour period, when the watch timing could also be measured against an atomic clock. I appreciate that declining mainspring torque may have an effect over the longer time period, but still think the result will be interesting.
A great video. If I ever get to the point where I start successfully putting watches back together I will certainly need one of these. Have you looked at any of the timegrapher telephone apps and tested any of them? I'm curious if any of them are worth examining
I've always suspected the Weishi was close but not accurate. I do find it useful to find relative positional variations or to gauge the effectiveness of a degaussing. I do NOT work on my watches, however.
I have a gmt movement eta2893-2 do you have to oil or grease the the 24 hour gear which is in a bag before you put it on the movement. I now it’s got nothing to do with weishi machine but if you can help it would be most appreciated.
Another great video! I'd have preferred to see three tests on each watch, on each machine for an average. And then done on at least one other watch. I think this would be a fairer comparison test.
I’ve been using an app as I get started. It’s ok but without a decent mic you can’t get enough signal gain for a more accurate analysis (had a good chat with the developer about this). That makes it ok for rate and beat error but useless for amplitude (unless it’s a loud movement). Time to move on up to a Weishi. What’s your take on a 1000 vs 1900 Alex?
Thank you for making this understandable. My timegrapher (got it few days ago) has a calibration port which looks like a USB-B connection. Is that something we can tune?
✨✨ Get Your Copy of my eBook, The Secrets for Adjusting Watch Rates ✨✨ ➡➡PDF DOWNLOAD watchrepairtutorials.com/ebook/ ⬅⬅ ➡➡Available on Kindle & Soft Cover @ geni.us/4x8C (Amazon) ✨✨Watchmaking Courses and Training ✨✨ ➡➡ watchrepairtutorials.com/watchmaker-course/ ⬅⬅
There are a couple minor differences. The Weishi 1000 has a monochrome LCD screen while the 1900 has a larger, high-resolution color display and has a slightly larger screen. Its a coin flip depending on what's important to you.
Thanks Alex, great video as usual. So if my Weishi shows my serviced watch is spot on I can brag about my skills and if it isn't I can blame Chinese quality.
if a watch is time calibrtaed on one of these but the lift angle was wrong ( set at 52 but should have been 50 degrees) would this show a reading ( secs /day ) that was good ( - 2 secs ) on the timegrapher but in reality would be much worse?
Hello I had a question is to see if there’s a difference between using a big jar or small jars for cleaning in my ultrasonic machines. I see that you always use big jars in your videos , I use 0.2 ounces do I need to use bigger ones
It depends on what you are going for. If you are just trying to see how accurate your watch is you don’t even need a timegrapher. Just check it against an atomic clock or quartz movement. That’s way more accurate in real life. Go answer your question as long as the timegrapher is picking up the sounds of the escapement it doesn’t matter if the strap is on or off.
Some equipment is for hobbies, and some equipment is for professionals. My equipment is ok for a $100 watch off eBay. If you lived near me, I would take my Omega, Tag, and Tudor to you. LOL Great videos. Thanks for sharing your knowledge.
Even if the timegrapher is not as accurate measuring amplitude, as long as it is consistent it should reflect any increases or decreases that you introduce to the movement, no?
Of course it will. My point was that if the amplitude in the machine and a incorrect lift angle is used a difference 25 degrees could make dialing a movement in a little more difficult.
There are test instruments for your more expensive time grapher to set it up from factory. Maybe its + or - .0000001 sec as pass and fail but all are very close to that initial calibration. This is way higher than the 1900 cheaper versions. Is like comparing a Rolls Royce to a Toyota Corolla. Both transport people and items so the end result is the same. For the 1900 you do a baseline test before any work to see how close you are to 0 secs and respectable amplitude. looking for a straight horizontal line. You are doing a good job if you achieve your goals using a 1900. It is a hobby and not trying to get absolute ultimate results. The watch runs better than before the rebuild.
On the Weishi I’m not sure or even if really matter one way or the other. It’s the screen data that’s important with this. On my etimer software it is the actually sound coming from the watch.
Lets consider this as an "unfair" test or comparison. Chinese timegrapher was off by several seconds againts the software timing machine. But how about the software? Is it accurate enough for timing solution or inaccurate enough compare to standard reference? When it comes to accuracy of the timer or timegrapher. The issue is how accurate or better to ask how stable the timebase is? Cheap timegrapher (in general) doesn't offer stability with temperature changes. Its not my opinion, it's a fact. Any timebase reacts to temperature changes and this changes affects the accuracy of the timer. Now, which of them telling us the "true" rate error of the watch? The software or the Chinese timegrapher?
Like everything in the Watchmaking arena there are a plethera of choices to make when it comes to these Timegraphers. Which is better the Weishi 1900 or the Weishi 1000? - I checked Amazon out and the prices range from the low 200's to 384 do know why that is as they all seem to be the same machine?
My understanding is that the only real difference is screen resolution, which is not that big of a deal so I would probably just go with the lowest price. I’m sure that most of them are made by one company and then they rebranded under different names.
@@watchrepairtutorialsI just asked the same question below Alex, so ignore that bit. Really interesting as I got a lot of encouragement on the subreddit to go 1900. Of course it’s not their money 😂
@@The_Scienceboy Hey man. I think the 1900 also has a function associated with picking up or inputting vibration rates. Im not positive as I never needed it. In that class of timing machines, again, i don't think it much matters. I have completely switched over to the eTimer which is a big jump up from the clone machines at around $400 plus mic and I just love it. It would not be as useful though if one doesn't know how to interpret all the Data that can be analyzed on the regular clone machine.
For the life of me (I´m not a pro, just a hobbyist) I have not seen an explanation on how to adjust to several positions. Do you measure first them (4, 5, 6, 8 whatever) and then average them out and adjust accordingly? If so, you need to adjust always (correct me if I'm wrong) in the dial down position which is the way you gain comfortable access to the regulating organ whatever it may be. So If you dial perfectly one position, I think the others will be thrown off...how is this accomplished? I dunno if this make sense, please help me, I haven't seen a video or a text going thoroughly with this topic.
Everything starts with the quality of the movement. This is not a simple answer that can be answered here. Keep watching. I have some video's on Dynamic Poising and Adjusting rates with the regulator pins coming up.
Thank you. I went through all the links on desktop and mobile and found that the newsletter link was messed up so i fixed that. All the other links in the description seemed to be fine. I even check incognito.
Have you ever checked out the mobile app phone timegraphers? As long as they get a clear enough signal from the mic I don't see why they would be any less accurate...
@@watchrepairtutorials Some of the apps show amplitude. I'm on android so idk what apps are available for iOS, I've been using 'Watch Accuracy Meter' which only shows beat rate and beat error, but I know 'Timegrapher' paid version shows amplitude. Not sure how accurate these apps are and some seem to work better than others, but they seem to work alright for pocket watches since the ticking is very loud. I haven't started servicing any watches yet so it's mostly just a heuristic. I imagine the only limiting factor would be the microphone signal since modern smartphones are extremely powerful, basically handheld computers...
@@zachary9685 I think for someone not into repair work that they are probably fine. But When you get into Positional timing, phone apps are going to be a pain in the ass.
So for the home hobbyist, the rate being off a couple seconds/day doesn't amount to much and that's okay. But the key is that if the Weishi amplitude was off by 15 degrees and the lift angle was off by 2 degrees and you don't want to make matters worse by guessing on the lift angle because everything adds up to 25 degree difference (or did you mean 17 degrees?), this level of inaccuracy won't help us diagnose amplitude problems. Sounds like that could be a frustrating, never-ending, rabbit hole for the home hobbyist! Given that, how would the home hobbyist then be able to diagnose amplitude issues with this level of inaccuracy? Not sure I want to purchase this machine? Any thoughts Alex? Thanks as always for your vids and expertise!
Perhaps the big takeaway here is to get the lift angle right. If a machine is off calibration by 5-10 degrees of amplitude by itself, that's not a huge problem, but add that to lift angle that is off by 3 , now your results OFF by 10-25 degrees and That is a problem when troubleshooting issues with rate.
@@watchrepairtutorialswhat I thought you’d say as I read the question. Guessing the LA would be like giving you the result to a different watch movement.
Citizen automatic ladies watch. Old purchased. The balance wheel running Too fast hours after adjusted. The. Lever, can’t. Control fast . And citizen watch. Crown can not. Released out please advise , how to control balance wheel , how to open crown. Key Rod Thanks
Sir I purchased old citizen automatic. Ladies. Watch , not cleaned, timing. Very fast. Hours in a day. Tried to adjust By. Plus and minus. Way. Lever, Then. Also. Timing fast can not ,reduced , and crown Key Rod cannot. Open , kindly advise As above. Appreciate your. Help thanks
@@raopenumaka3977 To correct the problems with this watch it needs a full service. It needs to be taken apart, inspected, all the parts that need to be cleaned, reassembled, and then lubricated. The problem could not be corrected until this is done .
Sir thanks for your prompts reply, citizen ladies small. Automatic. Cannot open. Key Rod. I. Think it is different way to open. Please. Give your valuable Advise. Thanks🙏
@@watchrepairtutorials I am using it, but I'm not a watchmaker and I don't have anything professional to check it against. I was curious how close is to real thing. wonderful channel by the way, glad I stumbled upon it!
@@OvidiuHretcanu it’s probably pretty close. You can always set the watch to an atomic clock and then track it over a 24 hour period of time as you wear it on your wrist. That’s really what’s important.
@@watchrepairtutorials yes indeed, for the time. I was curious about the beat error, which is not something you can double check. But then again, if I don't have the tools for that, it means also that I don't need to mees with it either :D (that video about magnetism was a savior for me. Managed to correct two pieces myself).
The problem with timegraphers is that who know what noise it will consider as part of the tick. For perfect accuracy and to be able to check for other noises one simply need to have to look on the sound pattern. In addidion to avoid wrong readings the intensity threshold stting is mandatory. Escapement analyzer if you can afford it or Watch-O-Scope for free with a simple piezo disk is perfect. If this is not a requirenment it makes no sense to buy any timegrapher. The watch apps are perfectly fine.
@@watchrepairtutorials I was looking but couldn’t see; did the professional software use the same LA as the Weishi? If so then of course the relative comparison remains valid.
You say the "Weishi" is inaccurate, but why do you assume the "Escapement Analyzer" is accurate? Also... the Weishi can be calibrated. I'm surprised they don't give you the tools to do that, but it has the USB port.
The Etimer escapement analyzer I'm using is calibrated to 1/10 of a second, so I can see how accurate it is. You can see this in the upper right of the screen as it's running. This is something that it is constantly doing so it maintains its accuracy. You are correct that the Weishi has a port on the rear which is used at the factory, but no watch supplier will calibrate this machine as they only will work on the Swiss brands.
@@watchrepairtutorials Now I wonder how accurate my own Weishi 1000 is. There's gotta be a way to measure that using an audio recording or oscilloscope for home tinkerers such as myself. It's a shame you can't get the software to calibrate it via USB.
The professional grade analyzer ; how did you "calibrate" that? You mentioned the error range for timing but not the other measurements. You're only as good as the calibrator that is doing the calibrating. And if your calibrator isn't calibrated properly....
isn't it better to just get the software which should be a lot more accurate for couple hundred more? TBH you probably could pirate it as well these days..
In today’s hectic and superficial world overcrowded with bogus experts this channel (you) is a real gem.
Old-school way ; proper, thorough, real-world explanation for both beginners and wannabe masters in the art of watchmaking.
No BS, right to the point, your presentation is involving and compelling.
I went through all of your lessons and I have no doubt that I will return to them many times to come.
My deep bow for will and effort to spread know-how.
Please, continue as long as you can. I trust that you have much more to share.
Wow brother thanks so much. This whole thing is keeping me pretty busy these days and I appreciate the heart felt words from you.
I got tons more coming. See you again.
Just a note of thanks for your well-structured and very clear videos. Like many I watch lots of U-tube material, and you are in a very small group of the very best, well qualified, patient, very generous and clear content posters. I, and I'm sure thousands of others, have gained hugely from your work and I'm massively grateful, thanks!
Wow thanks. That means a lot. 🥹
Yes, agreed. Thank you!!! As an amature, trying to learn as much as I can as a hobby, who knows what this could turn into. Thank you for your insights.
I have the exact same model and could not live without it, a Watch on the wrist can also tell a different story.
A timegrapher is a good indicator of how a movement is running for sure, they can also give a slightly different reading with an uncased movement.
Great video thank you.
Yes you are correct on all points. I didn’t take it out for this because it was just a head to head comparison.
I often find my watches on the wrist are performing much better than on the time grapher. On the other hand, there's a good deal of variation between the different positions that presumably compensate for each other.
I've also got that model.
You're exactly right. For hobbyist watchmakers who gain satisfaction from making an Ebay vintage come back to life and run well again, the Weishis are absolutely fine for gauging skill level and accomplishment. It's doubtful it ever happens, but if this hobby ever graduates to working on more expensive watches, I'll upgrade to a better timegrapher. In the meantime, my old Gruens and myriad ebauche watches will do just fine with a Weishi.
Thanks, Alex. Your videos are always thoughtful and well presented. I've learned tons from them.
Hey John. I appreciate you stopping by with the kind words. Keep at it brother.
I’ve found the Weishi timegrapher to be excellent for detecting strange noises that occur when something in the escapement is touching something that it shouldn’t. When the tracings are irregular or wild, it’s time for the microscope even if it’s keeping time in that position. It’s been a great tool to have!
Another great video, Alex!
Thanks as always Paul. I should have part 2 of Setting Pallet Stones out in a couple of days.
It's amazing that these tools exist for the modern watchmaker. I recently re-read "Longitutde" by Dava Sobel on the history of John Harrison and his quest for an accurate timekeeper to calculate Longitude for ships at sea. Mr. Harrison, I'm sure, would be amazed at the current state of mechanical movements and the tools available. Great video as usual. You are as great of an instructor as you are a watch maker. Thanks!
John, I am sure he would have been amazed but what about the guys from the early days of the Railroad watches that had to regulate watches in 5 positions. They would have loved it.
Thanks for the video so quickly Alex. 2spd is nothing to worry about for most watchmakers. Finding the lift angle on a clean movement and calibrating the 1900 with 180° on the balance was fun.
Right on Sir.
Great vid! This is almost identical to those of us in serious home theater calibration. The supplied mics with low and mid tier receivers are good enough to get things very close, but for highly accurate top level results you need the calibrated mics and the software that can use the calibration files.
Glad you enjoyed it! Like I said, for most people they are perfectly fine. I is nice to have the higher confidence level though
First: Thank you so much for your videos, I learned a lot with them and just started taking apart my first movements :-).
Just a little comment on this one, as I think you make quite an error with the comparison here:
You compare a one minute mean value of the escapement analyzer with a momentary value of the Wheishi (which is what is displayed big) so u compare two different things, especially as the value always has some scatter in it. If you take for example the mean of the last 4 12-second means of the Wheishi (+6 +6 +3 +3) you end up with a mean of 4,5 s/day, which is quite close to the mean of the escapement analyzer.
If you want to compare rates I think it´s good for both systems to have them sit for like 30 seconds and then compare the (mean) readings for the same time period. For the Weishi the default value for the means is 12 seconds, so take 5 of the smaller shown readings and calc. the mean.
Same is valid for amplitude, but there the Weishi only shows the momentary value, so it will be hard to compare
You are welcome.
I appreciate your input but let me make a couple points.
On timing machines, you don't consider the numbers from the "settling in" period. This would make your data wildly inaccurate and useless. Mean values are pulled from the last rate of the timing period.
Usually only the highest of quality movements can maintain a stable rate or amplitude for any length of time. Most will move up or down for various reasons related to the power transmission and is normal. WIldly fluctuating numbers are a signal of a fault and need correction before regulation.
There were two points to this video. The first is that these machines are close enough for a home watchmaker needs and the second point is that you don't want to guess as to what the lift angle is.
-Alex
The Weishi is a great hobby tool for the money. If you are attracted and inclined to the absorbing and myopic universe of watch tinkering being able to see your cleaning make an immediate difference is very satisfying. Great channel BTW.
Well said and 🙏
Great content as always Alex. I've been using a Weishi timergrapher (Chinese version) for a while, and it works very well.
Yea man they are pretty reliable all around.
would been lovely to see this against a phone app. thanks a lot for the video!
You're welcome! For repair any app or timing software that doesn’t measure the amplitude is pretty much useless.
It’s would be like trying to diagnose a quartz watch problem without testing the battery.
I have a Weishi 1000, I've done my best to test it by just obverving how accurately my watches (Planet Ocean 600m and Superocean 44) run over a few days compared to what the weishi is telling me and also by comparing its results with the watch accuracy meter phone app, everything seems to agree pretty well. I'm not a watchmaker or horologist I wouldn't open my Swiss watches and try to regulate or service them but I just find the Weishi handy to track how my watches are running and whether they need a service or not, I also have a few friends with watches who ask me to test theirs occasionally for the same reason. From the information I've read it seems the newer Weishis are a lot more accurate than older models they have improved the electronics and firmware, for instance an older weishi 1000 couldn't detect the amplitude of an Omega Coaxial movement properly but my newish 1000 seems to work properly in that regard.
It’s nice to hear about your experience with the Weishi timegrapher. Thanks for sharing!
remember too that there is software in both of those devices that will make them not completely comparable. I was a research scientist for years and whenever we compared data from different devices, we would frequently have to utilize another program to directly compare the results. For example, both of your datasets have 'smoothing' applied. That, in itself is an equation. What would be better is to open the movement, and utilize optical references to determine BA and AMP, and then run a multi-day test to judge accuracy, then compare back to both programs. I would wager that you'd then have 3 disagreements of data.
Well, pretty much everything you just said his way over my head so let me just say this.
This is a direct comparison of the ability of two separate sets of software, measuring the same thing. It’s barely to show the accuracy between one set of software that is highly calibrated versus a common Chinese variety timing machine.
Alex, I would love to see the sensor on your professional device in addition to the screen output. Is it a similar microphone set up?
Here's a picture of it. It has a USB added to it.
bit.ly/3j5bV2a
@@watchrepairtutorials that’s awesome! Thanks Alex!
I bought my red Swiss Witschi 25 year ago. Still working great.
I bet. What did it cost back then? $25.
Thanks again Alex, as always with measurement tools they need to be calibrated regularly but I don’t believe there is a facility available for the weishi timegraphers. I have experienced very expensive and regular requirements for calibration from my days in electronics many years ago. I don’t believe that with the tolerances needed in watchmaking it is too big a deal, the original railroad watches there were no timegraphers available to them.
Have to add, I also worked in electronics companies where calibration was less important and our repairs were fit for purpose. I think us amateurs to watchmaking can easily get too caught up with seeking perfection with a timegrapher rather than accepting that we have got it into specifications that a watch was originally designed for.
Totally agree Tom. There is a calibration port on the back but I don’t think I’ve ever heard of anyone calibrating one. I’m pretty sure that are primary used in the factory.
Thanks for another great video. In my experience the Weishi should generally be taken with a grain of salt, and perhaps even the expensive Swiss timing machines. Also in my experience the gain setting affects the amplitude reading quite a bit. To be sure I record a slo-mo video of the balance. Anyway, the Weishi is good enough to make sure things aren’t completely off.
Great point Its just a tool. Accuracy on the wrist is the bottom line. This was really for the guys who say just use 52 as a default LA. Its close enough. Well not when troubleshooting.
That's a good and honest comparison Alex. I recently bought the Weishi 1900 and it's a great little machine and is a confidence boosting guide that you are doing something right when comparing a pre and post service. Just finished servicing an ETA 2783 that indicated low amplitude on the Weishi. Post service it's running like a champ with good amplitude, beat error and rate. One question/observation - I do notice a difference in readings depending on whether the watch is cased compared to uncased or in a movement holder as compared to sitting directly on the stand. Do you think the microphone is affected by resonance from, for instance, a movement holder?
Great question. I am sure in certain extent but I’m no expert on the subject
Great video Alex! Curious how the Witschi machine would stack up against the Weishi -- basically if they're truly identical internally, or if the components in the Swiss version are better, or perhaps calibrated to a higher standard? That software you're using looks cool too, but how do you hook up the microphone?
The Swiss machine is going to be manufactured to higher standards being that’s it’s used professionally and it can also be calibrated. The Weishi has a port for calibration but I don’t personally know if any company calibrates them.
The I have a mic stand that plugs into the computer.
The Watch Escapement Analyzer and Timer version 2.172 looks interesting. Is the entry level $400 system what you use or do you use a different variant?
I use the etimer. It’s a fantastic system
It’s just as important to know precision (test/retest variability) in addition to accuracy. If precision is good the inaccuracy can be adjusted and the measurement will be reliable.
Thanks for clarifying this. 🧐
There was a you tuber who stopped posting once he went pro and he said when he got Omega certified , Omega told him to replace his weishi with a witchie 😮
Oh course. They are the industry standard.
Very good video as always Alex but I can't help thinking "how do you KNOW the expensive software is more accurate than the Weishi? I have a very low budget solution .. a small TBGC device from Ebay (about £30) communicated to by two open source software graphers both give similar but different readings to each other. Again, are any correct?
For what my opinion is worth I think they all give some indication of a running watch and are useful for seeing potential improvement post servicing
Sure they do. I am not saying they are bad by any means. The main take away might be to make sure the lift angle is correct.
Also my timer is calibrated and shows the deviation from the Atomic clock. It is calibrated to +/- .09 spd. You can see it in the bottom right of the screen
Thank you Alex. I have been looking at ebay avidly for a timegrapher. I think a 1000 would suit me, however would you consider it prudent to go for the 1900. Cost in the UK are £140 for the 1000 and £170 for the 1900. Makes a month extra saving for me. By the way, I have completed building my first two watches. Happy bunny.
Yes, absolutely. Ross, It's still a fine machine, so you can be confident in it. Happy Bunny
Excellent video - thanks. The very short test duration will also amplify any differences shown by the measurements. Can your calibrated software record and average readings over a longer time period as well as provide instantaneous readings? It would be interesting to see the test repeated over a 24 hour period, when the watch timing could also be measured against an atomic clock. I appreciate that declining mainspring torque may have an effect over the longer time period, but still think the result will be interesting.
Yes, the eTimer software can record rates/amplitude over a long period of time. Its a great way to test for isochronism.
A great video. If I ever get to the point where I start successfully putting watches back together I will certainly need one of these. Have you looked at any of the timegrapher telephone apps and tested any of them? I'm curious if any of them are worth examining
At the very eariy stages maybe but once you get into service work you need to know the amplitude and thats where they fall short.
@@watchrepairtutorials thanks. That's very useful to know!
Great video. What exectlly are you using for analysing?
The Etimer
Terrific video. Much appreciated. Thanks.
It’s always great to see you Sky Blue
Curious, will the model 1000 (or 1900), provide accurate results on a watch with a coaxial escapement? As a hobbyist should I care? Thanks!
As far as I know neither. You would need something like that Etimer.
I've always suspected the Weishi was close but not accurate. I do find it useful to find relative positional variations or to gauge the effectiveness of a degaussing. I do NOT work on my watches, however.
It’s close enough for the vast majority of people.
I have a gmt movement eta2893-2 do you have to oil or grease the the 24 hour gear which is in a bag before you put it on the movement. I now it’s got nothing to do with weishi machine but if you can help it would be most appreciated.
Yes, HP 1300 or D5 where it fits on the hr wheel.
Another great video! I'd have preferred to see three tests on each watch, on each machine for an average. And then done on at least one other watch. I think this would be a fairer comparison test.
Great point Andy. In hindsight I would have done more.
Perfect timing. Have thought about buying one 😀
You should!
I’ve been using an app as I get started. It’s ok but without a decent mic you can’t get enough signal gain for a more accurate analysis (had a good chat with the developer about this). That makes it ok for rate and beat error but useless for amplitude (unless it’s a loud movement).
Time to move on up to a Weishi.
What’s your take on a 1000 vs 1900 Alex?
See previous comment buddy. 👍
Excellent channel. I didn't hesitate to get your ebook.
Awesome, thank you!
Thank you for making this understandable. My timegrapher (got it few days ago) has a calibration port which looks like a USB-B connection. Is that something we can tune?
There are calibration services offered by suppliers but that would probably cost more than your machine. I don't know of a way to do it.
✨✨ Get Your Copy of my eBook, The Secrets for Adjusting Watch Rates ✨✨
➡➡PDF DOWNLOAD watchrepairtutorials.com/ebook/ ⬅⬅
➡➡Available on Kindle & Soft Cover @ geni.us/4x8C (Amazon)
✨✨Watchmaking Courses and Training ✨✨
➡➡ watchrepairtutorials.com/watchmaker-course/ ⬅⬅
Great learnings Alex!
Glad you think so Boyd.
Is the 1900 worth the small premium over the 1000? It seems the 1000 is more common. I'm trying to decide which to go for.
There are a couple minor differences. The Weishi 1000 has a monochrome LCD screen while the 1900 has a larger, high-resolution color display and has a slightly larger screen. Its a coin flip depending on what's important to you.
Thanks Alex, great video as usual. So if my Weishi shows my serviced watch is spot on I can brag about my skills and if it isn't I can blame Chinese quality.
Fair enough Mark
Interesting results. Thank you for the video.
Thanks for watching!
if a watch is time calibrtaed on one of these but the lift angle was wrong ( set at 52 but should have been 50 degrees) would this show a reading ( secs /day ) that was good ( - 2 secs ) on the timegrapher but in reality would be much worse?
No lift angle only affects amplitude readings not rate.
Helpful vid, thanks! I am noticing that your 1900 link currently goes to a 1000.
Thanks buddy. I fixed that.
Hello I had a question is to see if there’s a difference between using a big jar or small jars for cleaning in my ultrasonic machines. I see that you always use big jars in your videos , I use 0.2 ounces do I need to use bigger ones
If you are getting good results then there is no need to change.
Thank you fro getting back.
I have a watch with a 1 peice nato strap would I have to remove the strap so it can get accurate reading?
It depends on what you are going for.
If you are just trying to see how accurate your watch is you don’t even need a timegrapher. Just check it against an atomic clock or quartz movement. That’s way more accurate in real life.
Go answer your question as long as the timegrapher is picking up the sounds of the escapement it doesn’t matter if the strap is on or off.
Some equipment is for hobbies, and some equipment is for professionals. My equipment is ok for a $100 watch off eBay. If you lived near me, I would take my Omega, Tag, and Tudor to you. LOL Great videos. Thanks for sharing your knowledge.
You got it brother 🤙
Even if the timegrapher is not as accurate measuring amplitude, as long as it is consistent it should reflect any increases or decreases that you introduce to the movement, no?
Of course it will. My point was that if the amplitude in the machine and a incorrect lift angle is used a difference 25 degrees could make dialing a movement in a little more difficult.
I have a Weishi too, but the old model with the old lcd.
and how is it working for you?
@@watchrepairtutorials For my need is fine, but I do not have a top noch instrument to compare with
There are test instruments for your more expensive time grapher to set it up from factory. Maybe its + or - .0000001 sec as pass and fail but all are very close to that initial calibration. This is way higher than the 1900 cheaper versions. Is like comparing a Rolls Royce to a Toyota Corolla. Both transport people and items so the end result is the same. For the 1900 you do a baseline test before any work to see how close you are to 0 secs and respectable amplitude. looking for a straight horizontal line. You are doing a good job if you achieve your goals using a 1900. It is a hobby and not trying to get absolute ultimate results. The watch runs better than before the rebuild.
All good points. The purpose of this video was to answer the question. And to also make sure your lift angle is correct
I heard to escapement sound is a recording and not the actual sound of the watch. Any comment .Steve appreciated appreciated
On the Weishi I’m not sure or even if really matter one way or the other. It’s the screen data that’s important with this. On my etimer software it is the actually sound coming from the watch.
Lets consider this as an "unfair" test or comparison.
Chinese timegrapher was off by several seconds againts the software timing machine. But how about the software? Is it accurate enough for timing solution or inaccurate enough compare to standard reference?
When it comes to accuracy of the timer or timegrapher. The issue is how accurate or better to ask how stable the timebase is?
Cheap timegrapher (in general) doesn't offer stability with temperature changes. Its not my opinion, it's a fact. Any timebase reacts to temperature changes and this changes affects the accuracy of the timer. Now, which of them telling us the "true" rate error of the watch? The software or the Chinese timegrapher?
LOL
Like everything in the Watchmaking arena there are a plethera of choices to make when it comes to these Timegraphers. Which is better the Weishi 1900 or the Weishi 1000? - I checked Amazon out and the prices range from the low 200's to 384 do know why that is as they all seem to be the same machine?
My understanding is that the only real difference is screen resolution, which is not that big of a deal so I would probably just go with the lowest price. I’m sure that most of them are made by one company and then they rebranded under different names.
@@watchrepairtutorials Your honesty is refreshing Sir- Thank You Very Much
@@watchrepairtutorialsI just asked the same question below Alex, so ignore that bit. Really interesting as I got a lot of encouragement on the subreddit to go 1900. Of course it’s not their money 😂
@@The_Scienceboy Hey man. I think the 1900 also has a function associated with picking up or inputting vibration rates. Im not positive as I never needed it.
In that class of timing machines, again, i don't think it much matters.
I have completely switched over to the eTimer which is a big jump up from the clone machines at around $400 plus mic and I just love it.
It would not be as useful though if one doesn't know how to interpret all the Data that can be analyzed on the regular clone machine.
For the life of me (I´m not a pro, just a hobbyist) I have not seen an explanation on how to adjust to several positions. Do you measure first them (4, 5, 6, 8 whatever) and then average them out and adjust accordingly? If so, you need to adjust always (correct me if I'm wrong) in the dial down position which is the way you gain comfortable access to the regulating organ whatever it may be.
So If you dial perfectly one position, I think the others will be thrown off...how is this accomplished?
I dunno if this make sense, please help me, I haven't seen a video or a text going thoroughly with this topic.
Everything starts with the quality of the movement. This is not a simple answer that can be answered here. Keep watching. I have some video's on Dynamic Poising and Adjusting rates with the regulator pins coming up.
@@watchrepairtutorials Thanks so much! Love your channel!
Your Amazon links seem to be broken, clicking on them is producing an error page.
Thank you. I went through all the links on desktop and mobile and found that the newsletter link was messed up so i fixed that. All the other links in the description seemed to be fine. I even check incognito.
What those smaller numbers means??? Bellow the rate for example. Thank you
Those show the rate as it changes, while the movement is settling in to the position
Have you ever checked out the mobile app phone timegraphers? As long as they get a clear enough signal from the mic I don't see why they would be any less accurate...
What about amplitude?
@@watchrepairtutorials Some of the apps show amplitude. I'm on android so idk what apps are available for iOS, I've been using 'Watch Accuracy Meter' which only shows beat rate and beat error, but I know 'Timegrapher' paid version shows amplitude. Not sure how accurate these apps are and some seem to work better than others, but they seem to work alright for pocket watches since the ticking is very loud. I haven't started servicing any watches yet so it's mostly just a heuristic. I imagine the only limiting factor would be the microphone signal since modern smartphones are extremely powerful, basically handheld computers...
@@zachary9685 I think for someone not into repair work that they are probably fine. But When you get into Positional timing, phone apps are going to be a pain in the ass.
So for the home hobbyist, the rate being off a couple seconds/day doesn't amount to much and that's okay. But the key is that if the Weishi amplitude was off by 15 degrees and the lift angle was off by 2 degrees and you don't want to make matters worse by guessing on the lift angle because everything adds up to 25 degree difference (or did you mean 17 degrees?), this level of inaccuracy won't help us diagnose amplitude problems. Sounds like that could be a frustrating, never-ending, rabbit hole for the home hobbyist! Given that, how would the home hobbyist then be able to diagnose amplitude issues with this level of inaccuracy? Not sure I want to purchase this machine? Any thoughts Alex? Thanks as always for your vids and expertise!
Perhaps the big takeaway here is to get the lift angle right. If a machine is off calibration by 5-10 degrees of amplitude by itself, that's not a huge problem, but add that to lift angle that is off by 3 , now your results OFF by 10-25 degrees and That is a problem when troubleshooting issues with rate.
@@watchrepairtutorialswhat I thought you’d say as I read the question. Guessing the LA would be like giving you the result to a different watch movement.
@@The_Scienceboy Exactly. makes it hard to diagnose problems when you don't have the right information in front of you.
There is phone app called watch accuracy meter on android phone
Can you please test it and let us know if we can rely on it .. thank you so much
I use an Iphone and I don't see it in their store.
Citizen automatic ladies watch. Old purchased. The balance wheel running
Too fast hours after adjusted. The. Lever, can’t. Control fast . And citizen watch. Crown can not. Released out please advise , how to control balance wheel , how to open crown. Key Rod
Thanks
I'd like to offer some advice, but honestly, I don't know what you're asking. Has this watch been serviced properly?
Sir I purchased old citizen automatic. Ladies. Watch , not cleaned, timing. Very fast. Hours in a day. Tried to adjust
By. Plus and minus. Way. Lever, Then. Also. Timing fast can not ,reduced , and crown Key Rod cannot. Open , kindly advise
As above. Appreciate your. Help thanks
@@raopenumaka3977 To correct the problems with this watch it needs a full service. It needs to be taken apart, inspected, all the parts that need to be cleaned, reassembled, and then lubricated.
The problem could not be corrected until this is done .
Sir thanks for your prompts reply, citizen ladies small. Automatic. Cannot open. Key Rod. I. Think it is different way to open. Please. Give your valuable
Advise. Thanks🙏
@@raopenumaka3977 more than likely it is a snap on back. Look around the edge of the back for a notch to put your blade in to pop it open
how about Watch Accuracy Meter Google Play app? how far is this one?
I don’t know. I use a iPhone. Let me know if you test it.
@@watchrepairtutorials I am using it, but I'm not a watchmaker and I don't have anything professional to check it against. I was curious how close is to real thing. wonderful channel by the way, glad I stumbled upon it!
@@OvidiuHretcanu it’s probably pretty close. You can always set the watch to an atomic clock and then track it over a 24 hour period of time as you wear it on your wrist. That’s really what’s important.
@@watchrepairtutorials yes indeed, for the time. I was curious about the beat error, which is not something you can double check. But then again, if I don't have the tools for that, it means also that I don't need to mees with it either :D (that video about magnetism was a savior for me. Managed to correct two pieces myself).
Very helpful thanks
No problem. Thanks Stevie
The problem with timegraphers is that who know what noise it will consider as part of the tick. For perfect accuracy and to be able to check for other noises one simply need to have to look on the sound pattern. In addidion to avoid wrong readings the intensity threshold stting is mandatory. Escapement analyzer if you can afford it or Watch-O-Scope for free with a simple piezo disk is perfect. If this is not a requirenment it makes no sense to buy any timegrapher. The watch apps are perfectly fine.
So you’re not a fan of of timegraphers but are a fan of phone apps.
Ok
@@watchrepairtutorials Not exactly but ok ;)
BOUGHT MINE BUT THE DISPLAY ONLY LASTED A FEW WEEKS , ITS VERY FAINT NOW AND HARDLY USEABLE ,
Sounds like you got a bad one. Was there a particular brand
JUST SAYS N0, 1000@@watchrepairtutorials
The lift angle of that movement is 50 degrees not 52. The Weishi will give the wrong results unless the lift angle is set correctly.
It really doesn’t matter in a side by side comparison does it now.
@@watchrepairtutorials I was looking but couldn’t see; did the professional software use the same LA as the Weishi? If so then of course the relative comparison remains valid.
@@youregrammersucks yep, you can see it on the right side of the tool bar
You say the "Weishi" is inaccurate, but why do you assume the "Escapement Analyzer" is accurate?
Also... the Weishi can be calibrated. I'm surprised they don't give you the tools to do that, but it has the USB port.
The Etimer escapement analyzer I'm using is calibrated to 1/10 of a second, so I can see how accurate it is. You can see this in the upper right of the screen as it's running.
This is something that it is constantly doing so it maintains its accuracy.
You are correct that the Weishi has a port on the rear which is used at the factory, but no watch supplier will calibrate this machine as they only will work on the Swiss brands.
@@watchrepairtutorials Now I wonder how accurate my own Weishi 1000 is. There's gotta be a way to measure that using an audio recording or oscilloscope for home tinkerers such as myself. It's a shame you can't get the software to calibrate it via USB.
@@peetiegonzalez1845 oh I agree. I'm sure that there are some smart electronics guys out there that could figure it out.
The professional grade analyzer ; how did you "calibrate" that? You mentioned the error range for timing but not the other measurements.
You're only as good as the calibrator that is doing the calibrating. And if your calibrator isn't calibrated properly....
It is calibrated through the Simple Network Time Protocol (SNTP).
@@watchrepairtutorials So that calibrates amplitude and other items besides time?
@@rockymntdan1 No, it calibrates the software .
isn't it better to just get the software which should be a lot more accurate for couple hundred more? TBH you probably could pirate it as well these days..
For most people it’s accurate enough. As with anything else and watch repair, you have to look at the return of value for the money you’re spending.
👍🤝🆒
So far see you don't know how to use a timgrapher or how to read it.
I’m sorry what are you saying?