Follow-up about my Neoden K1830 issues and Neoden Customer Support

[seon]

Hey folks,

I wanted to do a follow-up about my issues with my K1830 ( aka Neoden 8 ) PnP and the lack of support I was getting from Neoden with regards to replacing faulty/failing components.

Neoden did finally step up and send me replacement lighting modules for both of my UP cameras (front and back) and a new comms controller board to replace the one in my machine that has missing (blown off caps). They also sent some replacement solenoids.

I've now managed to get my PnP up and running again - the replacement lighting modules did fix the camera issues I've been having.

I'm still having some air related issues with 5/8 of the nozzle pneumatics - but I'm going to track that down myself.

It was super frustrating to have them take this long to agree to support their hardware, but they did in the end - which is important, so I wanted to share for those of you still thinking about purchasing a Neoden product.

I want to also add that EMLogic were instrumental in getting all of this to happen, and in getting the parts I needed identified and shipped to me.

Cheers

Seon
Unexpected Maker
 

[Kean]

Good idea to post this update.  Thanks Seon.

I'm now going to see if NeoDen can help me with adjusting the pick up positions on my TM240A.  I need a password for this apparently...
Positions are slightly out of whack and seem to be getting worse, so it will often fail to pick or sometimes reject 0603 parts due to vacuum leakage, and of course if it does pick the part then any rotation is off centre.

[48X24X48X]

But if you were to keep silence and never mentioned them in bits and pieces on this forum, do you think they would actually step up? Obviously they are scared to be a 2nd Charmhigh. Simple as that. That's purely my opinion though.

[Kean]

But if you were to keep silence and never mentioned them in bits and pieces on this forum, do you think they would actually step up? Obviously they are scared to be a 2nd Charmhigh. Simple as that. That's purely my opinion though.

I agree 100%.  They knew about the problems for a long time.

Ongoing support is critical for these machines.  The documentation is quite basic, and many parts are near impossible to source.

My needs are simple, so I'm OK to suffer with my second hand NeoDen and Seon's old Charmhigh... 

[Reckless]

It sounds pretty bad to have these kind of issues.  Machines from reputable manufacturers don't have these kind of problems and cost the same in used market.  I was expecting much more both in reliability and customer service from neoden.  Neoden 8 caught my interest being super lightweight at 250kg, most of the machines I work with are 1600-1800kg.

How long have you had it?  How many hours have you run on it?  Single shift?

[loki42]

I think you're in a unique position to have tried so many brands of machines.  I think 90% of the people with a Chinese machine haven't ever run a non Chinese machine. Most probably have never seen one.  Even people with reputable brands have mostly only ever run the brand they have so they don't know if their experience is a normal level of challenge.  I chatted with one person who had worked for Bosch and run big lines of Fuji,  uic, Panasonic and ASM / Siemens which was useful but uncommon.  Your Fuji will have put down more parts per feeder than these have per machine but I'm not sure it's realistic to compare,  the Fuji requires a different level of care and feeding with regards to power,  air, size.  Were some of the machines you bought second hand sticky difficult to get working?  Some people aren't used to the concept of used stuff and are worried they'll get something they can't get working. 

My MPM was a nightmare to get working.  Had a faulty servo controller,  fluff stuck on a scale sensor,  broken input sensor,  broken input air filter,  dodgy camera cable,  door sensors that needed repair.  It took a lot of time to fix all that.  The machine now works well and hopefully ready for a few million more prints but I can see why someone might consider new.

[Reckless]

I have had more bad luck with chinese equipment than used equipment.  I have had brand new chinese machines not do what they said they were supposed to.  It has happened more than 5 times.  So thats why I can't stand chinese stuff.  It just doesn't do what its supposed to - poor engineering, poor design.  I have had bad luck once or twice with a used name brand machine but overall there is better support and a light at the end of the tunnel.  The chinese machines I bought never worked for what they were intended and were not designed to work ever.  It was an expensive pretty box.  With name brand equipment even if they need a repair its once and forget about it.  It might be a costly repair but then you get the machine back to new/like new.  Fuji is famous for being built to last forever so thats why I love their equipment.

In the end the issue is more about what the machine was bought for.  Some people prefer something that doesn't last a year, and others want 10+ years.  I love my MPMs but I bought DOS based ones.  I have had a few minor repairs around $50 for various cards that popped during brown outs.  I did have a fiber optic camera cable I had to replace which ran $125.

My expectation from Neoden is that they are trying to be a top class Chinese SMT manufacture.  On this I am rooting for them hard and want to see they deliver quality, reliable products.  I would love to buy one of their machines over a Fuji but they need to make dependable machines.  I really like their small neoden 9, I consider it a competitor to an Essemtec Fox.  I also liked Neoden8 for its lightweight until I read this thread. 

[seon]

Hardware fails - it's a fact of life... Do Chinese made machines tend to fail more often than their more expensive counterparts? Probably, and it's not just because of poorer construction and inferior parts - it's mostly because the companies never truely test their machines/design - never "run them in" in real world scenarios - test them for 1000's of hours on a line, doing actual assembly, do the things that USERS do to validate the machines. Do things badly to see how they cope, stretch them, abuse them, use them beyond their own expectations for the machines.

Until they start doing that, the "real world" quality level and reliability will never improve. 

I cant afford new non "Chinese made machines" - and I'm just not prepared to buy/import a second hand machine I've no experience with, and then have to deal with other new and unexpected issues that all entails. The key here is to go in eyes wide open when buying Chinese machines.

I've decided I'll continue to use Neoden machines (I'm putting in a second N8 on my line to eliminate my PnP as single point of failure) - and I'm putting in a new Neoden IN12C (using my current IN6 as a fallback to eliminate that single point of failure) and from now on I'll treat them as a disposable item - Buy a machine and expect it to only last a year, and that's what I budget into my production line up keep. Sounds crazy, but at the prices of these machines, it's still way cheaper than buying alternative "non Chinese" machines. If the machine lasts 18 month or 2 years, I'm even further ahead.

Still dreaming of buying a Juki RS-1R though ... but it's just a dream!

Seon
Unexpected Maker

[Mangozac]

Seon is right - I've said all along it's horses for courses.

Our 6 head Kayo machine has served us really well for the past 2.5 years. It was brand new and came with a warranty (we never had any breakdowns so I don't know what the warranty is worth) and was the perfect option to bootstrap our SMT line and learn the process. Kayo support was fantastic, including plenty of software and feature updates along the way. The few bugs we had were sorted quite quickly. I did a lot of evaluation before buying, even visiting China, and I'm glad I chose the Kayo over Neoden (at that time anyway, I haven't compared current offerings).

We've now reached a point where our volume and requirements can justify moving up to a pair of secondhand Yamaha YG series machines. Our first was installed a couple of weeks ago and we're still commissioning (fixing a few leaking air cylinders, etc.) but we're very impressed. These are amazing, highly capable machines but we've found actually take a lot longer to learn how to use. This is because there are so many more settings and complex things to understand. Don't get me wrong; once we are fully trained the Yamahas will be so much quicker to program and use. Not to mention the vastly higher placement speed.

Sounds crazy, but at the prices of these machines, it's still way cheaper than buying alternative "non Chinese" machines. If the machine lasts 18 month or 2 years, I'm even further ahead.

Once you are familiar with the ins and outs of pick and place machines and if you don't mind tinkering I'd recommend looking at secondhand "Western" options. My original plan was to run one Yamaha and the Kayo (since they both use CL feeders), but a second Yamaha came up at a price I couldn't refuse so the Kayo will be moving on. These machines are built like tanks, designed to run 24/7 for years. At the puny few thousand hours per year most of them do in small to medium CMs they will last forever so are an awesome option for those of us with smaller businesses and the skills to maintain them.

it's mostly because the companies never truely test their machines/design - never "run them in" in real world scenarios - test them for 1000's of hours on a line, doing actual assembly, do the things that USERS do to validate the machines. Do things badly to see how they cope, stretch them, abuse them, use them beyond their own expectations for the machines.

I agree with that assessment.

[loki42]

I think that's the thing with any mainstream machine.  Fuji uses Fuji machines to build Fuji machines.  They are designed to be run 24x7 at insane placement rates. My uic machines were very simple to get setup and have been very heavily used and very poorly maintained for the last 15 years. Lubricating and levelling was 90% of the time oh and reading the docs.... My stencil printer has done millions of prints.  It was an arse to get working but if I spent a few K to get the local very friendly distributor (Suba) to fix it instead of fixing it myself I'd still have an excellent bit of kit for very little money.  I wasted time and money fixing it myself. 

I think Chinese aoi / spi might be worth looking into as Chinese factories seem to actually use some Chinese machines in that area. When I read reviews of Chinese cnc gear it's always oh it's usable once you replace all this stuff and fix all that stuff.  It's a fun project and good content for YouTube but if you want a lathe for manufacturing you buy a used one or a new mainstream brand. 

[Reckless]

Its nice to have varying opinions/view points on this board.  We all have different backgrounds/industries so its expected we will have differences in philosophies.  We are all after the same thing cheap, works well and reliable.  I think one of the main differences is that on our side of the world chinese machines cost alot for freight and 25% duty which makes it very expensive.  Generally even more than used reputable manufacturers in US.  I am rooting for china to make competitive products to bring prices down.  I feel manufacturers are charging too much for same machine and prices need to come down considerably.  One thing I also like about China is they think outside the box and offer mechanical feeders which are much cheaper than electronic.  It seems like all manufacturers collectively came together to stop offering mechanical feeders. 

I'm also hoping China starts making reliable industrial ovens, Essemtec has stopped making them since they were acquired.  They offer service/support for 10 years but they were my goto oven company as it was a no-nonsense reliable machine.  The only item I like buying from reputable companies is the screen printer only because that is where 80% of errors come from.   I'm waiting for china to learn jetting as I like that technology to replace stencil printers.  I believe china knows contact dispensing but not jetting.  Unfortunately contact dispensing is too slow to replace a stencil printer.

But seeing this thread and others on eevblog has made me lose respect for neoden as a company.  It seems like they are still producing inferior products and not learning from their mistakes.  My expectation is that they would learn and improve their products in each progressive generation. 

[48X24X48X]

I think Neoden started making larger machine not very long ago, probably around 5 or less years. Before that it was a failure product S1 that Seon too had a unit before it was returned. It was pulled out of the market soon after it was launched due to numerous problems and issues. Before that they were selling the N4 which was their very first attempt on vision system, conveyor belt and PC based controller. I believe they they try very hard to be Essemtec (S1 is the living proof) to fulfill the overseas market but at the same time needed the China market support which actually doesn't really need these smaller and slower machines despite more "advance" as what they claimed. Generally outside of China, people would only know Neoden or Charmhigh due to their marketing. But believe me, there many more better and much older companies that has been in this industry for much longer with far more advanced machine. Some of them are not even sold outside of China. These are what small and medium factories in China are using to churn out bulk of electronics you see in your country. If you do your homework enough, you will get what you need at the price point you willing to pay. I personally using a HWGC machine and I think they are great from support wise and hardware quality. Just like Mongozac with his Kayo machine, there's little to non issues based on our experience. And HWGC actually uses their own machine to build their own electronics too. Even one of their show room in SZ is actually a PCBA factory that I have visited back in 2019. So I guess, generalizing all machine coming out from China has bad quality, bad support and so forth would be a little misguiding and false at times.

[Yellofriend]

Machines from reputable manufacturers don't have these kind of problems and cost the same in used market. 

You are probably right. And they are build by people that actually know what they are doing. But this said, when you a problem there, and you will eventually, you pay way, way more for support (because it is available) and for parts.

Best is maybe (aka cheapest) convert to OpenPnP and solve every future problem yourself. You just need a controller for at least 5 axle (if you have dual nozzle). Controller will cost maybe $70

[seon]

I think Neoden started making larger machine not very long ago, probably around 5 or less years. Before that it was a failure product S1 that Seon too had a unit before it was returned. It was pulled out of the market soon after it was launched due to numerous problems and issues. Before that they were selling the N4 which was their very first attempt on vision system, conveyor belt and PC based controller. I believe they they try very hard to be Essemtec (S1 is the living proof) to fulfill the overseas market but at the same time needed the China market support which actually doesn't really need these smaller and slower machines despite more "advance" as what they claimed. Generally outside of China, people would only know Neoden or Charmhigh due to their marketing. But believe me, there many more better and much older companies that has been in this industry for much longer with far more advanced machine. Some of them are not even sold outside of China. These are what small and medium factories in China are using to churn out bulk of electronics you see in your country. If you do your homework enough, you will get what you need at the price point you willing to pay. I personally using a HWGC machine and I think they are great from support wise and hardware quality. Just like Mongozac with his Kayo machine, there's little to non issues based on our experience. And HWGC actually uses their own machine to build their own electronics too. Even one of their show room in SZ is actually a PCBA factory that I have visited back in 2019. So I guess, generalizing all machine coming out from China has bad quality, bad support and so forth would be a little misguiding and false at times.

No, Neoden had the N7 well before the S1 ever shipped and tanked. The N7 is the N8 (K1830) but only pneumatic feeders, and uses on the fly cameras on the nozzles instead of the front and back up cameras. The N8 was their second large machine and came years after the N7 was out in the wild and being hammered by many happy users.  Still, yes, they are relatively new in this space compared to other brands - but are not dong the "clone everyone else's design" like  most other Chinese companies do.

They had the perfect opportunity to set a new high bar for everyone else to catch up to and follow, but they have struggled to change the "usual Chinese way of thinking and doing business" that is the key to being successful outside of China. I stil hope that it wil happen - one day!

Seon
Unexpected Maker

[Reckless]

Hey folks,

I wanted to do a follow-up about my issues with my K1830 ( aka Neoden 8 ) PnP and the lack of support I was getting from Neoden with regards to replacing faulty/failing components.

Neoden did finally step up and send me replacement lighting modules for both of my UP cameras (front and back) and a new comms controller board to replace the one in my machine that has missing (blown off caps). They also sent some replacement solenoids.

I've now managed to get my PnP up and running again - the replacement lighting modules did fix the camera issues I've been having.

I'm still having some air related issues with 5/8 of the nozzle pneumatics - but I'm going to track that down myself.

It was super frustrating to have them take this long to agree to support their hardware, but they did in the end - which is important, so I wanted to share for those of you still thinking about purchasing a Neoden product.

I want to also add that EMLogic were instrumental in getting all of this to happen, and in getting the parts I needed identified and shipped to me.

Cheers

Seon
Unexpected Maker

Any update since?  Feels strange that soo many parts failed so fast.  I really like the neoden 9 form factor but hearing these problems has scared me away.

[sinewave]

Seon is moving away from the K1830, I believe he has been looking into purchasing a Juki. The K1830 I've been working with hasn't seemed to have many issues. I've learnt quite well with NeoDen machines that if you want to do proper production and not expect issues, don't buy a Neoden and just get a used older Yamaha or similar.

[PCBA_MAKER]

I'm going to have to agree with all that is said here.  After 4 months struggling with the Neoden 10 I have to say that I'm very disappointed. Although I haven't experienced hardware failure I am experiencing component drift both lateral and rotational. 

[Reckless]

I'm going to have to agree with all that is said here.  After 4 months struggling with the Neoden 10 I have to say that I'm very disappointed. Although I haven't experienced hardware failure I am experiencing component drift both lateral and rotational.

It sounds like a algorithm issue.  How many boards have you run?  Is it something you can recalibrate to fix? 

[seon]

Still dreaming of buying a Juki RS-1R though ... but it's just a dream!

Seon
Unexpected Maker

Turns out it wasn't a dream after all - and I'm now the proud owner of a new Juki RS-1R!

You just never know what the future will bring

Seon
Unexpected Maker

[Reckless]

Still dreaming of buying a Juki RS-1R though ... but it's just a dream!

Seon
Unexpected Maker

Turns out it wasn't a dream after all - and I'm now the proud owner of a new Juki RS-1R!

You just never know what the future will bring

Seon
Unexpected Maker

Congrats!  Going from Neoden to Juki, it's like going from the bottom to the top.  I used to have older KE-760/750/740/730s and they were super slow at 0402.  I also had 570 models which I could never get to work but had nice form factor.  Someone convinced me it was the laser system and to go with cameras only so I got rid of my 8 Jukis for 6 GSMs.  I have been more interested in Yamaha over Juki lately but have heard how reliable Juki has become.  How come your usb connectors are not pushed down properly?  Also why is it running so slow, I would have thought new Jukis are IPC rated at 26k cph.  You are only getting 12k.

[seon]

Still dreaming of buying a Juki RS-1R though ... but it's just a dream!

Seon
Unexpected Maker

Turns out it wasn't a dream after all - and I'm now the proud owner of a new Juki RS-1R!

You just never know what the future will bring

Seon
Unexpected Maker

Congrats!  Going from Neoden to Juki, it's like going from the bottom to the top.  I used to have older KE-760/750/740/730s and they were super slow at 0402.  I also had 570 models which I could never get to work but had nice form factor.  Someone convinced me it was the laser system and to go with cameras only so I got rid of my 8 Jukis for 6 GSMs.  I have been more interested in Yamaha over Juki lately but have heard how reliable Juki has become.  How come your usb connectors are not pushed down properly?  Also why is it running so slow, I would have thought new Jukis are IPC rated at 26k cph.  You are only getting 12k.

The RS-1R is IPC rated for 30K cph, but you don't get that when you don't allow the system to tell you were to place feeders for optimal performance per job

I'm sharing a static feeder layout across all of my products so I never have to move any feeder ever - but that means sacrificing performance.

I also only have 2 of each nozzle size above the 7500's so the machine can't optimise that either right now,. but I have more nozzles on the way.

I'm also not allowing it to place across circuits atm - so it also can't optimise there. I want to be 100% confident before I go all out on allowing the machine to go nuts.

Even with all of those limitations put on the optimiser, I'm still assembling boards 4-6 times faster than the Neoden did, with 100% accuracy  I'm pretty happy with that for now

Seon
Unexpected Maker

[Reckless]

In the video you said neoden was doing it in 6-8 minutes.  4:43 doesn't sound that much improvement?  Is there a reason Juki can't place usb connectors properly?  I never done one just curious what issue is faced.  Not having enough nozzles can definitely have a huge impact on cph.  I never cared for feeder optimization other than on parts that go on repeatedly.  You are still learning the machine so getting 33% rated performance is expected, but my expectation would be to get 70% IPC rating.

[seon]

In the video I said the Neoden did it in over 16mins.  Not 6-8.  So that’s over 3.x faster so far.
Re the USB. I didn’t compensate for the peg leg height, so wasn’t pushing the connector down far enough. A user error, not a Juki thing 🤪

Seon
Unexpected Maker

[Reckless]

Thanks for the clarification, I haven't seen Chinese PnP equipment.  Good to hear user error not Juki error.