TWS Automation Quadra DVC Evo - someone owns this machine? (pick and place)

[macck]

Hello everyone,
Im looking for a pick and place machine for company i work with. We are producing low volume series of PCB's and we want to automate it.
I found distributor in Poland (renex), and they have this TWS machine and also Neoden and Novastar machines.

We are looking for maybe not too fast machine because this is not the case, hovewer pretty accurate and with soldering paste dispenser as we want to reduce human work on assembly to minimum.

Hard to say about price tag. Lets say $15,000 for pick and place, $5000 for reflow oven (novastar?). Maybe more, but lets try to stick to this price tag for now. We are not interested about buing machine in china, and so we are not interested in used equipment.

Me personally, i like this TWS machine very much. Its perfectly sized for our needs, and pretty fast for what we do, and also might be equipped with paste dispenser. Hovewer I'm unable to find any reviews about this system. Just this video from manufacturer i guess on youtube



I know there are some of you working everyday with automatic production lines. Can you suggest me something?

Best  regards and have a great day!

[mrpackethead]

Unless you are prepared to pay upwards of $100k, forget paste dispense for somethign that will be useful for anythign smaller than 0603, and 1.27 mm pitch soic.
If you dont' want to Chinese, or 2nd hand, you're going to be in the order of $80-90k minimum.

[macck]

So lets say, we are abandoning this paste dispense idea and stick to stencil's. Let's even forget the price for now.
Please take a look at this offer http://www.renex.info/pokaz_kategorie.php?cid=54 - is any of these machines familliar and can anybody tell me anything about how they perform?

[SMTech]

Unless you are prepared to pay upwards of $100k, forget paste dispense for somethign that will be useful for anythign smaller than 0603, and 1.27 mm pitch soic.
If you dont' want to Chinese, or 2nd hand, you're going to be in the order of $80-90k minimum.

Forget paste dispenser full stop, they sound like an awesome solution but unless you have a very specific need the speed is poor, the maintenance high and performance below that of a stencil. Typically dispensing options on P&P are for glue or adding extra paste to specific pads which will only be needed for highly atypical designs. Essemtec & Mycronic make the only realistic paste dispensing options that can come close to replacing an automated stencil printer.

You won't find reviews online for any of these machines, to choose you need to work out what you need, go and see the machines in action and talk in person to existing customers. The distributor should be able to help you with that.

Yamaha/IPulse are excellent mid to high end machines, you won't find anyone complaining about owning one but they are not cheap (not expensive compared to similar machines either). In 2012 in the UK an M20 was £95KGBP+feeders and you could get a decent starter package including feeders for ~£125k GBP. However that was 2012 and these machines are Japanese which means the cost can change a lot with the exchange rate & inflation.

TWS & DDM have been selling machines for quite some time, before the Chinese came along you could probably have considered them the entry level although arguably the non-chinese entry level is the Polish Mechatronika these days. You will find reviews on those as the basic model is quite cheap; they seem to be affordable but slow. There is a series of english videos where someone sets up and runs some PCBs, it looked very clunky to me but seemed to do the job.

The best mix of price, build quality and performance IMO is probably Autotronik (http://www.labem.pl/Autotronik-automaty.htm) built in China, designed & supported from Germany. You should be able to get one of those with quite a few feeders for well under $100K, in the UK in 2011/12 it was more like 60-70K. Essemtec machines are some level above this these days, what you get with Essemtec is much slicker software that is very easy to use but you do start getting quite close to the cost of faster more accurate machines that may have other advantages.

You should also buy a printer, there are some interesting semi-automatic ones these days (e.g Miniko or the TWS SR2700) that address some of the repeatability concerns you might face doing it by hand without the expense of an automated one. There's also PBT https://www.pbt-works.com/smt-printers/semi-automatic another eastern european manufacturer with a good reputation in this area. However big fully automatic printers are built like tanks & it is well worth considering buying 2nd user, just make sure the machine is still supported and you know where to find a service engineer for it.

[mikeselectricstuff]

Hard to say about price tag. Lets say $15,000 for pick and place,

We are not interested about buing machine in china, and so we are not interested in used equipment.

In that case you're out of luck.
For that budget, China or old used are your only options.

[macck]

As I mentioned, forget about this price. I just found one new Neo D4 machine for around $15k and assumed that for the size this might be the range.

[SMTech]

The Chinese machines are MUCH cheaper so you can't use them as a guide, they are also about 20 years behind. Everyone else doesn't really publish their prices, they will tailor their quote to your request and apply some mysterious discounts, feeder pricing is particularly meaningless as they get a lot cheaper when purchased in bulk with a machine. The one exception to this I am aware is http://www.intelligentdrives.com/ who have their pricing right there for all to see.

In the US Manncorp also gives some indicative pricing if you log in, or did...

[mikeselectricstuff]

The Chinese machines are MUCH cheaper so you can't use them as a guide, they are also about 20 years behind.

They are also targeted at a market where labour is cheap, so features that help reduce the need for babysitting tend to be lower priority than making it cheap.
AFAIK no Chinese P&P has the ability to deal sensibly with feeder problems by carrying on with other feeders until it's placed everything it can before asking for assistance.

[rx8pilot]

Hard to say about price tag. Lets say $15,000 for pick and place, $5000 for reflow oven (novastar?). Maybe more, but lets try to stick to this price tag for now. We are not interested about buing machine in china, and so we are not interested in used equipment.

The numbers are not well aligned with your demands.

For reference - I have a modestly capable machine that was built in the 90's that I purchased broken and fixed myself. I have about $12k in it plus an ENORMOUS amount of blood, sweat, and tears fixing, calibrating, and learning to use it on my own. If I purchased it ready to go from a used dealer or the OEM - it would be 3x-4x the cost. This is for an OLD machine. The next option in that price range is Neoden - which is new but considerably more limited in capability and capacity. A DDM Novastar machine can be had for about your budget(I think), but it is very limited - small and slow designed for prototypes. Kind of something that is more of an 'assembly helper' than a real pick and place solution.

Feeders and accessories balloon the cost of most systems. More than half the money spent on my system are feeders and I got them all used off of eBay.

Also - consider the challenge of managing expectations. If the boss writes a $20k check and expects PCB's to be flying out of the machine after a few days - he/she will likely be very disappointed. P&P has endless surprises ready to punch you in the gut at any moment. There always seems to be some problem, fault, or limitation that slows you down or stops work altogether. A fiddly feeder, a software glitch, a part that wont fit, vacuum pump overheated, etc, etc, etc......

My personal experience with P&P is doing low-volume high mix assembly on customer demand. Overall - I am very happy I have the system but it has taken a few years off my life to get there. The remaining life left on my old machine is relatively limited and I am already doing research for it's replacement.

[mrpackethead]

Forget paste dispenser full stop, they sound like an awesome solution but unless you have a very specific need the speed is poor, the maintenance high and performance below that of a stencil. Typically dispensing options on P&P are for glue or adding extra paste to specific pads which will only be needed for highly atypical designs. Essemtec & Mycronic make the only realistic paste dispensing options that can come close to replacing an automated stencil printer.

MY-500,  about $400k USD.  QUite impressive.    theres a crowed in texas using one, they do lots of one off prototype assmebly work. Makes a lot of sence for them,    But would be super slow for anything more than a few boards.

Yamaha/IPulse are excellent mid to high end machines, you won't find anyone complaining about owning one but they are not cheap (not expensive compared to similar machines either). In 2012 in the UK an M20 was £95KGBP+feeders and you could get a decent starter package including feeders for ~£125k GBP. However that was 2012 and these machines are Japanese which means the cost can change a lot with the exchange rate & inflation.

yeah,  I'd add Juki to that.  I'm a yamaha shop,    You can get some older 2nd hand machines.. a MUCh better option than buying new low quality ones.. ( they will just give you endless trouble ).  Be prepared to invest plenty of time to learn...

I can point you in teh direction of a great company in China who sells 2nd hand refurbed Yamaha machines.   http://www.ksunsmt.com/   I did'tn buy my orginal machiens from them, but thye have been wonderful in providing parts, feeders, and knowhow to keep my machines running.   I'm very likely to go and buy another 2nd hand machine from them next year.    A 20 year old YV-100ii will eat a Neoden4 any day of the week. 

You should also buy a printer, there are some interesting semi-automatic ones these days (e.g Miniko or the TWS SR2700) that address some of the repeatability concerns you might face doing it by hand without the expense of an automated one. There's also PBT https://www.pbt-works.com/smt-printers/semi-automatic another eastern european manufacturer with a good reputation in this area. However big fully automatic printers are built like tanks & it is well worth considering buying 2nd user, just make sure the machine is still supported and you know where to find a service engineer for it.

Yes, echo that comment.   A fully automated printer will cost you a bunch and wont' gain you much over a decent manual or semi-automatic will. Again be prepared to spend a LOT of time getting this part of your process right, or you'll have endless trouble.

[mrpackethead]

All that said, 20,000 UKP + a lot of work can get you somethign workable.   But you'll need to be super prepared to work really hard, and do lots of homework.

[SMTech]

Nice auction of used machines https://www.go-dove.com/en/events?cmd=details&event=561414& the AX501/301 wouldn't suit but some of the others might. Showed up in my email this morning & apparently ends 2morrow.. not exactly heavily bid on.

[SMTech]

In the UK here is a 2nd user TWS http://www.shawline.co.uk/product.php?id_product=157, save you signing up £9500 no feeders, they reckon it was ~£45k new. If they do work outside the UK (they certainly bring machines in) I do know they are a bit more than a plain broker, deliver to the factory floor, set you up etc. In a way holding 1/4 of its value after 10 years might be a bit expensive, but these smaller machines seem harder to find.

[macck]

Thank you for your feedback.

It gave me a lot of information. I allready talked about the costs, but our funding source for such thnigs are projects etc, so we are limited about vendors and buying second hand. I send SMTech post to my boss and he just asked "But Yamaha/IPulse will do?" so I guess more or less if we gonna buy something, it's gonna be machine that does the job.  As i said, these 15k was my idea after seing Neoden. Nothing more   I worked already on some small pick and place machine, however i dont remember the brand. Nevertheless it introduced me to some really basic problems. And our production volume isn't that big to be worried about speed. We could do more productive stuff instead of soldering hundreds of 0603's by hand.

[mrpackethead]

If you are going to buy a line of Yamaha kit, by the time you have all your feeders + all the other bits that go along with, it, you'll need to have upwards of 200,000

[jmelson]

Well, I have a better report than RX8pilot.  I bought a used Philips CSM84 machine when it was about 12 years old (1995 built, I purchased in 2007).  The prior owner threw in about 50 feeders and some other stuff.  I had it running in a week, having NEVER seen a P&P before.  (I did do a lot of homework before the purchase.)  The CSM84 is a somewhat old-school machine that uses mechanical chuck jaws on the nozzles to center the small parts.  It has an alignment station to center the large parts (SOW and FPGA type).  But, it is capable of handling a wide range of part types.  If you only have one machine, then you need that.  (It can handle parts up to 6.5 mm tall, that is a bit of a limitation.)

I have had a few issues over the last decade I have run it for low-volume use.  A conveyor sensor became intermittent, and it had to get bad enough so that I could be sure what, exactly was bad.  The worst was the rotation motor got its commutator filled with copper dust and the machine was shutting down.  Also some hoses cracked, and every once in a while I have to tear down the venturi vacuum generators for cleaning.  And, of course, always fooling with balky feeders and component tapes with WILDLY varying adhesion of the cover tape.  But, mostly, it just chugs along making boards.

My machine does not have vision, and that seems to be something that could help it be more accurate.  So, if I ever replace it, that will be something I have to have.  Pretty standard nowadays.

Unless you are going to be making boards several hours per day, every week, I can't imagine justifying the expense of a new major-maker machine in the $100K++ range.  There are a LOT of used machines on the market with reasonable prices, especially if you DON'T go through a broker.

Jon

[rx8pilot]

Well, I have a better report than RX8pilot.  I bought a used Philips CSM84 machine when it was about 12 years old (1995 built, I purchased in 2007). 

Curious how much it cost all in when you purchased it.

If I had the money - I would have NOT chosen the fixer upper machine. It was a huge risk and a huge time commitment. At the time - it maxed out my financial resources. So glad I was eventually able to make it work.

[jmelson]

Well, I have a better report than RX8pilot.  I bought a used Philips CSM84 machine when it was about 12 years old (1995 built, I purchased in 2007). 

Curious how much it cost all in when you purchased it.

If I had the money - I would have NOT chosen the fixer upper machine. It was a huge risk and a huge time commitment. At the time - it maxed out my financial resources. So glad I was eventually able to make it work.

OK, the seller was "motivated", they had 2 brand-new machines coming in a week, and needed the old ones out.
Every time we exchanged calls or emails, the deal got sweeter.  I think I paid either $3300 or $3600 for the machine to be palletized and brought down out of the 2nd story to be ready to load on a truck.  At first, I was just going to get the bare machine, but he then threw in some feeders, then it was ALL his feeders, which ran to about 50 units.
He also threw in the electronic guts of a CSM60 which was from a different vintage, so nothing was compatible with my machine.  I traded that electronic control box to another guy for a bunch of feeders in the 12 and 16 mm size, which I didn't have quite enough of.  I also got several sets of Philips books for the several machines he had there.
So, that was parts, maintenance, operation, etc.

A few months later he sent me a homemade feeder alignment base and the feeder mounting bars out of the old machine.  These have been quite useful.

I realize not too many people are going to get deals like this on what was pretty close to a fully functioning machine.  The only thing that was off was the calibration of the alignment station and rear feeder were zeroed out.

Yes, I know, I lucked out, sight unseen, I could have gotten a total junker from this guy.  Or, had something really nasty happen to it on the truck.

Jon

[rx8pilot]

Yes, I know, I lucked out, sight unseen, I could have gotten a total junker from this guy.  Or, had something really nasty happen to it on the truck.

Jon

Oh...yeah!!! That is amazing.....and rare. 

[SMTech]

There are a couple of significant obstacles to buying 2nd user machines.
If not buying from a broker/dealer it can require a bit of luck to find the machine you are after when you need it, middlemen might add cost but they can also reduce risk.
Even quite successful companies tend not to have $$$ of cash sitting in their bank accounts for capital expenditure and even if they did it isn't always tax efficient or wise to spend it all on capital equipment.
Finance companies and Government grant schemes often do not support the purchase of 2nd user equipment. (they might however allow for factory refurb to "as new").

The OP is looking for a machine as a step-up from manually soldering 0603, pretty much anything is a step up from that. New machines are a big investment and its important to choose the machine that fits the way you work as well as how you might work in the future. Having a machine like that will alter your designs as you realise what you can do and exploit its potential further, however while Yamaha Machines are excellent, I think they are perhaps a step to far to start with, plus depending on what your designs look like you might even end up wanting two to get the feeder space.  I would look at the smaller machines first, you can do things with standalone machines like the TWS (or Autotronik or Essemtec or ID or Fritsch if configured without a conveyor) that big in-line machine cannot. That can include putting trays (including strips of tape) in the placement area with the PCB without losing feeder space, loading & building multiple designs at once & managing boards that the designer has forgotten important things like borders and fiducials on.

I would look at Fritsch, Autotronik, Heeb?,Essemtec and TWS, first they will cost you less and take up less space.
Moving up a little another strong machine is the Europlacer IICO, the flexibility on these machines is pretty much unrivalled, it's no speed demon but you could keep a sizeable proportion of your stock on up 198 lanes ready go on the machine at the drop of a hat.
Next port of call might be the Europlacer IIneo or Mycronic MY300, however I'm not sure exactly what the pricing on these looks like when you get a real quote on these anymore as they have both had recent updates and my old quotes had BIG discounts on them.
After that I think you end up in your Yamaha territory, there have been a lot of mergers so there less to choose from than there were but you still have Juki (now includes sony), Hanwha (samsung), Kulicke&Soffa (assembleon), Mirae, ASM (Siemens/DEK) & Fuji.

K&S I think you can ignore, ASM have the "E" series, the rest you can ignore. Everyone else has a range of machines all of which should be comparable in price to the Yamahas where I would be looking at the Ipulse M10/20. I would look at all of them at least at the brochure level, you can learn a lot just from the features they choose to highlight.

Of course I have left out two US brands Universal make some awesome looking machines and then there is Versatec who might sell something called a C5.

[mikeselectricstuff]

Chinese machines are now getting so cheap that for a first step from manual assembly, it may actually be worth buying one as a learning experience, to inform about what you may want for a higher-end system.

[mrpackethead]

theres a good project going on with one of those charming tech machines wth openpnp.    I have had some expereince with teh QiHe TVM machiens, but i'd probably suggest avoiding them now.

An old yamaha or Juki will eat any of these new low cost machines,  There is no substitue for just building it like a tank.

[Nauris]

I would look at Fritsch, Autotronik, Heeb?,Essemtec and TWS, first they will cost you less and take up less space.
Moving up a little another strong machine is the Europlacer IICO, the flexibility on these machines is pretty much unrivalled, it's no speed demon but you could keep a sizeable proportion of your stock on up 198 lanes ready go on the machine at the drop of a hat.
Next port of call might be the Europlacer IIneo or Mycronic MY300, however I'm not sure exactly what the pricing on these looks like when you get a real quote on these anymore as they have both had recent updates and my old quotes had BIG discounts on them.
After that I think you end up in your Yamaha territory, there have been a lot of mergers so there less to choose from than there were but you still have Juki (now includes sony), Hanwha (samsung), Kulicke&Soffa (assembleon), Mirae, ASM (Siemens/DEK) & Fuji.

K&S I think you can ignore, ASM have the "E" series, the rest you can ignore. Everyone else has a range of machines all of which should be comparable in price to the Yamahas where I would be looking at the Ipulse M10/20. I would look at all of them at least at the brochure level, you can learn a lot just from the features they choose to highlight.

Of course I have left out two US brands Universal make some awesome looking machines and then there is Versatec who might sell something called a C5.

Then there is also MIMOT Gmbh. Their MB-series have up to 324 feeder lanes, apparently very flexible and made specially for small production runs. I don't know but got the impression that their machines may be quite expensive.

[SMTech]

Yeah Mimot seem to target a similar market to Europlacer/Mycronic but without the reach, seem very German speaking nation focused.

[jmelson]

theres a good project going on with one of those charming tech machines wth openpnp.    I have had some expereince with teh QiHe TVM machiens, but i'd probably suggest avoiding them now.

An old yamaha or Juki will eat any of these new low cost machines,  There is no substitue for just building it like a tank.

Yes, but it is not ONLY the hardware.  I have been reading about OpenPNP, as well as the new Chinese machines, and apparently, there are a lot of software issues.

My Philips CSM84 has quite solid software.  There were a few error messages that were not in the book, or the book didn't really explain what the real problem was.  There was one that had me going for a while, "data overflow".  Well, the machine accumulates errors and statistics for EVERY board that has a program loaded, for every head and feeder.  If that fills the available space, the machine will stop with the "data overflow" error.  I eventually figured out to clear the machine statistics after every build was complete, no more problem.  But, I had to guess what the error actually meant.  My problem was I never cleared that data.

But, I've never had any problems with the machine placing parts in the wrong place when the build program was correct, and my machine diesn't have vision, so that eliminates a whole range of errors right there.

These machines (Philips/Assembleon, Juki, Mydata, Quad, etc.) were all meant to be run by people with a modest level of training, and to just keep chugging along making boards.  They are designed to recover from as many failures (mispicks, etc.) as possible and continue running.

Jon