NeoDen YY1 Pick And Place Machine With Under $3K Price for Hobbiest/Low vol Usag

[EEVblog]

Yes, for something like this, slow is perfectly fine.  But if you spend so much time having to muck around with the machine, work around its shortcomings, and wasting components due to it constantly screwing up...  Suddenly its not worth bothering with anymore.
Striking a balance between "I'll just assemble it by hand" and "I'll just outsource it to someone with a real machine" kinda demands a level of fuss that's below what this machine seems to have.

The sole purpose of a PnP machine is to automate the task of placing components. Regardless of how good a PnP machine you have, there is a fixed setup time which is roughly similar for all classes of machines (pre-loaded feeders aside), the advantage only comes in reliability where you can press go and walk away.
The more you have to massage the machine the more of a time waste it becomes, to the point of uselessness eventually.
A slow 100% reliable machine will beat a super fast expensive machine that has constant issues, every time.

[loki42]

If you need a machine that just works then Suba have a nice used Juki for $20k with very low hours.  It's an older machine (not as old as my machines) but actually works.  If you want something that's good for YouTube content improving, fixing and repairing it then the Chinese ones are probably a better choice. 

[asmi]

The sole purpose of a PnP machine is to automate the task of placing components. Regardless of how good a PnP machine you have, there is a fixed setup time which is roughly similar for all classes of machines (pre-loaded feeders aside), the advantage only comes in reliability where you can press go and walk away.

The bold part is an important distinction between machines designed for high-volume production, and those designed for prototyping. The latter compromise on feeders, and are designed to be loaded with reels once and never touched again (unless a reel runs out), with parts distinct for a specific design delivered via ad-hoc feeding like loose parts, strip holders, etc.

A slow 100% reliable machine will beat a super fast expensive machine that has constant issues, every time.

The problem is - it's not gonna be a 3K$ machine, but more like 30K$ one. That's the way this market seems to work. And no machine is ever 100% reliable, even super-expensive ones fail every once in a while.

[asmi]

If you need a machine that just works then Suba have a nice used Juki for $20k with very low hours.  It's an older machine (not as old as my machines) but actually works.

You probably forgot to mention that is weighs like a car and has a size similar to that of a car, while here we're talking about desktop machine. Desk-top, not garage-floor-top.

[newto]

If you need a machine that just works then Suba have a nice used Juki for $20k with very low hours.  It's an older machine (not as old as my machines) but actually works.

You probably forgot to mention that is weighs like a car and has a size similar to that of a car, while here we're talking about desktop machine. Desk-top, not garage-floor-top.

Yep, I'm still happy with our machine, even with the teething problems. We have very low production volume (30-60 boards a year) of a single product, and we were spending over 100 CAD per board to have them assembled (and management specifically forbids outsourcing assembly out of the country). No way I could get management to approve a 20k used system (plus whatever it would cost to deliver) that would take up even more of the floorspace we don't have.

The biggest problem we're running into right now is that our old version of Mentor PADS didn't put solder mask in 0.2mm pad gaps, so we keep getting bridges on a bunch of our components, and I'd never noticed before because the assembly shop never complained.

Next step is a complete redesign, for easier to acquire parts, better layouts, better solder mask, and to share as many components as possible with a second product I'm working on.

This is not the machine for someone doing 10k panels a year, or multiple product lines that can't all fit in one machine.

I may try to come up with a design of some kind of lock that keeps the tape in place when not being advanced by the pin, or some kind of mechanism that is triggered by the pin, but always advances exactly the right spacing.

[tboicey]

As a random datapoint, I make about 1000/boards a year on my Neoden 3V without trying too hard. Some of my higher volume stuff I order fully assembled from AllPCB, but the stuff that we need 100 or so of here and there are done on the Neoden. It's just too much work to deal with a third party for small volumes and especially with evolving products.

The most I ever recall doing in one session is probably around 75 boards. That board had a somewhat low component count but that's most of a day just feeding the machines.

I just wanted to throw that out there because I'm seeing the phrase "prototyping or high volume" and I suspect I'm really somewhere in between.

[jmelson]

Yep, I'm still happy with our machine, even with the teething problems. We have very low production volume (30-60 boards a year) of a single product, and we were spending over 100 CAD per board to have them assembled (and management specifically forbids outsourcing assembly out of the country). No way I could get management to approve a 20k used system (plus whatever it would cost to deliver) that would take up even more of the floorspace we don't have.

The biggest problem we're running into right now is that our old version of Mentor PADS didn't put solder mask in 0.2mm pad gaps, so we keep getting bridges on a bunch of our components, and I'd never noticed before because the assembly shop never complained.

Next step is a complete redesign, for easier to acquire parts, better layouts, better solder mask, and to share as many components as possible with a second product I'm working on.

This is not the machine for someone doing 10k panels a year, or multiple product lines that can't all fit in one machine.

I may try to come up with a design of some kind of lock that keeps the tape in place when not being advanced by the pin, or some kind of mechanism that is triggered by the pin, but always advances exactly the right spacing.

I have about $7500 US invested in a Quad/Samsung machine from an auction.  That includes shipping, feeders, repair parts, etc. 
As for bridges, good solder paste is key, we use GC10, which is totally fantastic!  Also, reducing stencil aperture size is very important.  The smaller the lead pitch, the more you have to shrink the apertures.  I do down to 0.5mm pitch FPGAs.  Finally, accruate alignment of the stencil and part placement is required to eliminate bridges.
Jon

[newto]

I have about $7500 US invested in a Quad/Samsung machine from an auction.  That includes shipping, feeders, repair parts, etc. 
As for bridges, good solder paste is key, we use GC10, which is totally fantastic!  Also, reducing stencil aperture size is very important.  The smaller the lead pitch, the more you have to shrink the apertures.  I do down to 0.5mm pitch FPGAs.  Finally, accruate alignment of the stencil and part placement is required to eliminate bridges.
Jon

There were probably more cost efficient ways of getting a pick and place, but my team is 5 people including me, and only two of us are actually hardware/assembly related, and neither of us has experience troubleshooting a PnP, so I wanted something where I wouldn't need to worry about something complex breaking.

Solder wise, we're using Chip Quik SAC305, which looks like a similar product, but I'll try to get the GC10 for the next batch (last time I looked, it was hard to get in Canada, but that may have changed. I've also modified the design files to make a version with smaller apertures on the troublesome parts, I just haven't gotten around to ordering a new stencil. Almost certainly will help though; for whatever reason, two of the components are the same package, but one has much smaller pads on the footprint, and we haven't that part bridge a single time, while the larger footprint component bridges almost every time.

[jmelson]

I have about $7500 US invested in a Quad/Samsung machine from an auction.  That includes shipping, feeders, repair parts, etc. 
As for bridges, good solder paste is key, we use GC10, which is totally fantastic!  Also, reducing stencil aperture size is very important.  The smaller the lead pitch, the more you have to shrink the apertures.  I do down to 0.5mm pitch FPGAs.  Finally, accruate alignment of the stencil and part placement is required to eliminate bridges.
Jon

There were probably more cost efficient ways of getting a pick and place, but my team is 5 people including me, and only two of us are actually hardware/assembly related, and neither of us has experience troubleshooting a PnP, so I wanted something where I wouldn't need to worry about something complex breaking.

Solder wise, we're using Chip Quik SAC305, which looks like a similar product, but I'll try to get the GC10 for the next batch (last time I looked, it was hard to get in Canada, but that may have changed. I've also modified the design files to make a version with smaller apertures on the troublesome parts, I just haven't gotten around to ordering a new stencil. Almost certainly will help though; for whatever reason, two of the components are the same package, but one has much smaller pads on the footprint, and we haven't that part bridge a single time, while the larger footprint component bridges almost every time.

My first P&P came out of a running shop that was upgrading.  It ran with only two SMALL issues for 13 years.  If I had KNOWN what level of disaster the newer machine would be, I would not have bought it!  It had apparently sat in unconditioned space in Austin, TX for over 6 years, and had been hacked on by monkeys, and I'm giving my simian cousins a bad name there!  When I got it delivered, it ran for 3 days, and then the servo system would not come out of E-stop anymore.  I spent WAY too long trying to scientifically diagnose the fault before buying a replacement board.  Then, I found out a motor had been replaced with the wrong type, and that in the process a ball spline had lost all the balls.  Then, a number of boards started failing on it, requiring expensive replacements.  Fortunately, this was a very popular model in the far East, so there are lots of brokers with affordable spare parts.  No component I have replaced has failed again so far!  So, if the machine had been properly stored and not tinkered with by idiots, I would have likely spent about $4000 to get it moved and running - mostly just feeders.
Jon

[EEVblog]

If you need a machine that just works then Suba have a nice used Juki for $20k with very low hours.  It's an older machine (not as old as my machines) but actually works.  If you want something that's good for YouTube content improving, fixing and repairing it then the Chinese ones are probably a better choice.

When the YY1 came out I was thing that it would make great to see what type fo small scale manufacturing is possible with a low cost machine (say sub $5k). I don't want to have to be doing videos (and spending time) dicking around troubleshooting the thing. I wanted "here's a low cost machine, let's make some boards and see what the limitations are". Not a 10 part series on how I finally got it to work reliably.
Maybe that would be popular, I don't know, but it's not something that interests me.

[48X24X48X]

I used to have a push feeder with integrated feeder designed in China sold by German guy for about $4K excluding an external PC. I came from manual picking by hand for years and was churning out thousands of boards just with a tweezer and hacked toaster oven. I thought this $4K machine would scale up my speed but I was dead wrong. I wasted more time trying to get the machine to work and was pulling my hair out until I decided I have to get rid of the machine. A gentlemen bought it from me and he eventually got fed up of it too. Bought myself an $8K machine with some of that cost goes into the removable Yamaha feeders and shipping fee and never look back. I don't have to play with all these silly feeder issues or funny software implementation. I guess if you have a little bit more extra than the $5K, skip these smaller desktop machine as you might eventually out grown the capacity, or you get something needing too much of baby sitting, or it is just not working reliable enough to replace your years of lightning speed of manual hand placing.

[EEVblog]

I guess if you have a little bit more extra than the $5K, skip these smaller desktop machine as you might eventually out grown the capacity, or you get something needing too much of baby sitting, or it is just not working reliable enough to replace your years of lightning speed of manual hand placing.

That's the thing, I'm looking for the (inevitable?) "game changer" PnP machine that is affordable, reliable, small, and realtively capable. I thought the YY1 was it, but many reports say otherwise.

[mikeselectricstuff]

I guess if you have a little bit more extra than the $5K, skip these smaller desktop machine as you might eventually out grown the capacity, or you get something needing too much of baby sitting, or it is just not working reliable enough to replace your years of lightning speed of manual hand placing.

That's the thing, I'm looking for the (inevitable?) "game changer" PnP machine that is affordable, reliable, small, and realtively capable. I thought the YY1 was it, but many reports say otherwise.

IMO The Opulo LumenPnP is worth keeping an eye on - I'm sure it's going to take time to get it right, they're only on their first iteration of feeders, but having spoken at length to the people behind it, I think it has a good chance of becoming a useful machine.
They actually use it in-house to assemble their own products, which is where a lot of companies like Neoden fall down, without using a machine day-to-day, you have little chance of finding all the issues that real users will face.
Anothr issue with larger companies is they don't want a low-end machine to compete with their higher-end products.
 

[EEVblog]

I guess if you have a little bit more extra than the $5K, skip these smaller desktop machine as you might eventually out grown the capacity, or you get something needing too much of baby sitting, or it is just not working reliable enough to replace your years of lightning speed of manual hand placing.

That's the thing, I'm looking for the (inevitable?) "game changer" PnP machine that is affordable, reliable, small, and realtively capable. I thought the YY1 was it, but many reports say otherwise.

IMO The Opulo LumenPnP is worth keeping an eye on - I'm sure it's going to take time to get it right, they're only on their first iteration of feeders, but having spoken at length to the people behind it, I think it has a good chance of becoming a useful machine.

Agreed, but until it's available a commercial product you can just buy and use, it doesn't fit the criteria of the type of video series I wanted to do.

[loki42]

I guess if you have a little bit more extra than the $5K, skip these smaller desktop machine as you might eventually out grown the capacity, or you get something needing too much of baby sitting, or it is just not working reliable enough to replace your years of lightning speed of manual hand placing.

That's the thing, I'm looking for the (inevitable?) "game changer" PnP machine that is affordable, reliable, small, and realtively capable. I thought the YY1 was it, but many reports say otherwise.

IMO The Opulo LumenPnP is worth keeping an eye on - I'm sure it's going to take time to get it right, they're only on their first iteration of feeders, but having spoken at length to the people behind it, I think it has a good chance of becoming a useful machine.
They actually use it in-house to assemble their own products, which is where a lot of companies like Neoden fall down, without using a machine day-to-day, you have little chance of finding all the issues that real users will face.
Anothr issue with larger companies is they don't want a low-end machine to compete with their higher-end products.

Have the Opulo people used other mainstream brands?  I think it's handy to know what level of annoyance is "normal". With a lot of the hobby gear in shocked they often haven't looked at any other gear and are starting from scratch.  It seems really common in cnc routers that people haven't worked somewhere with a mainstream machine or done a short course or anything on one.  I got into a free government sponsored machining course for CNC stuff but it was run through a local maker space and they had a really basic Roland desktop mill as the educational machine... I'm sure a tafe would have at least had a manual machine as well and some experience.

[Kjelt]

That's the thing, I'm looking for the (inevitable?) "game changer" PnP machine that is affordable, reliable, small, and realtively capable. 

What is your definition for affordable defined in US$ ?
What is your definition of reliable in #failed placeings/performed placings ?

As many have pointed out the main challenging issues are feeders and software.
The basis for a PnP machine is no rocket science any electronics engineer can build the powercabinet with powersupplies motor drivers cabling etc.
The only challenge is alignment which for a PnP machine that gets no load to endure can be post calibrated if needed.

So I wonder why there are not more PnP kits from Ali with an interface to OpenPNP ?
Even you Dave step a bit out your electronics domain into the mechatronics domain and a new world of opportunities opens up for you, or do you not think that people would like to watch a 20 episode DIY build of a pnp machine ?

[MakeIt]

The problem of the feeders is also the problem when using openpnp.

The feeders will be a big part of the price when you want a good functioning system.

I tried it with a 3D printer and some strips. What I see when having this setup is that the tape can shift a bit when it is hit too hard with the nozzle.
this is part of the problem. The system needs to be able to pick up without shifting the tape. This means I need to go back to the drawing board.

Talking to a guy in Germany on the YY1 subject, he found this same problem with the YY1. He told me, he had "bent" the pin after doing some trials.
I am not sure if it comes from the tape or if he hit something else.

[coppice]

I guess if you have a little bit more extra than the $5K, skip these smaller desktop machine as you might eventually out grown the capacity, or you get something needing too much of baby sitting, or it is just not working reliable enough to replace your years of lightning speed of manual hand placing.

That's the thing, I'm looking for the (inevitable?) "game changer" PnP machine that is affordable, reliable, small, and realtively capable. I thought the YY1 was it, but many reports say otherwise.

IMO The Opulo LumenPnP is worth keeping an eye on - I'm sure it's going to take time to get it right, they're only on their first iteration of feeders, but having spoken at length to the people behind it, I think it has a good chance of becoming a useful machine.
They actually use it in-house to assemble their own products, which is where a lot of companies like Neoden fall down, without using a machine day-to-day, you have little chance of finding all the issues that real users will face.
Anothr issue with larger companies is they don't want a low-end machine to compete with their higher-end products.

Those things look interesting from their web site. Have you seen one working, know anything about its accuracy, etc.?

[coppice]

I guess if you have a little bit more extra than the $5K, skip these smaller desktop machine as you might eventually out grown the capacity, or you get something needing too much of baby sitting, or it is just not working reliable enough to replace your years of lightning speed of manual hand placing.

That's the thing, I'm looking for the (inevitable?) "game changer" PnP machine that is affordable, reliable, small, and realtively capable. I thought the YY1 was it, but many reports say otherwise.

IMO The Opulo LumenPnP is worth keeping an eye on - I'm sure it's going to take time to get it right, they're only on their first iteration of feeders, but having spoken at length to the people behind it, I think it has a good chance of becoming a useful machine.

Agreed, but until it's available a commercial product you can just buy and use, it doesn't fit the criteria of the type of video series I wanted to do.

They claim the new one (V3) needs only very quick and simple assembly, and you are up an running. They say minutes rather than hours or days, but it takes more than minutes to read the instructions for most things, so I assume that is an exaggeration. Still, it might be worth a look.

[EEVblog]

They claim the new one (V3) needs only very quick and simple assembly, and you are up an running. They say minutes rather than hours or days, but it takes more than minutes to read the instructions for most things, so I assume that is an exaggeration. Still, it might be worth a look.

Yes, and powered feeders coming this quarter but not available just yet. It's getting there.

[mikeselectricstuff]

I have seen a prototype Opulo feeder in person, though not actually running. Bit hazy on the details now but recall it had some good & bad points, and will definitely need at least another couple of iterations, and they understand that.
 I did give them some input on the sort of things they need to test for, particularly as many different tape types as they can - reliable peel-off.

[sinewave]

I guess if you have a little bit more extra than the $5K, skip these smaller desktop machine as you might eventually out grown the capacity, or you get something needing too much of baby sitting, or it is just not working reliable enough to replace your years of lightning speed of manual hand placing.

That's the thing, I'm looking for the (inevitable?) "game changer" PnP machine that is affordable, reliable, small, and realtively capable. I thought the YY1 was it, but many reports say otherwise.

IMO The Opulo LumenPnP is worth keeping an eye on - I'm sure it's going to take time to get it right, they're only on their first iteration of feeders, but having spoken at length to the people behind it, I think it has a good chance of becoming a useful machine.

Agreed, but until it's available a commercial product you can just buy and use, it doesn't fit the criteria of the type of video series I wanted to do.

They claim the new one (V3) needs only very quick and simple assembly, and you are up an running. They say minutes rather than hours or days, but it takes more than minutes to read the instructions for most things, so I assume that is an exaggeration. Still, it might be worth a look.

I had the YY1 assembling in probably 30 or so minutes, only had to read the manual to figure out the password for the parameters display as calibration was needed. The YY1 is your best choice for low-volume production, and I'm not a NeoDen shill, though I have not tried any other cheap PnPs.

[bugrobotics]

Took the plunge and ordered one.  I'll document and update with my experience.

[level6]

I've had mine for a week now and I'm quite satisfied with the purchase. Looking forward to your impressions.

[itsrealfast]

I'm thinking of buying one. Please let me know your thoughts. Also I dont mind buying a used one if anybody upgraded

Thanks