Skip to main content
Topic: SMT700 New vision based SMT machine from SMALLSMT (Read 74006 times) previous topic - next topic

SMT700 New vision based SMT machine from SMALLSMT

I am designing a small vision based machine having 60x 8mm feeders exchangeable at a size of 800x700mm.
We meet all CE requirements and the EU machine directive.
The production will start end of year.
    
Please note that all pictures are wider than shown you get the complete size on right mouse click.

Here are some pictures of the construction 50% of design task finished.










Re: New vision based SMT machine

Reply #1
picture 5, what is happening with the tape that is coming into the machine here?, do the operator need to sit and snip it off when it's coming in?, I cant see anything forcing it down or away in some way?

Re: New vision based SMT machine

Reply #2
No the design is not finished yet there are some details missing.
The feeder design needs to be completed.
We solved 50% the design task is still going on.
But you see a compact machine having a lots of feeders and using vision system to align the components.

 

Re: SMT700 New vision based SMT machine

Reply #3
Quote
Looks great! Care to share any spec's, approximate pricing, etc?

We plan to release the machine in different versions.

The starter version has stepper drives using leadshine DSP drivers (quiet servo like) and 3 feeder cassettes no cover
starting at 5499USD.
For the EU machine directive you need the cover and security equipment (lock switch, security controller) add 550USD.
A full featured machine using 6 feeder cassettes and cover will start at 6750USD.
The machine is controlled by a PC software having several import functions for different cad file formats.
We could do fiducial recognition to align the board and component alignment using different correction methods.

I will post more informations in the next weeks.

Re: SMT700 New vision based SMT machine

Reply #4
how many reels would the different models support?

Re: SMT700 New vision based SMT machine

Reply #5
The SMT700 machine supports 6 slots.
You could use 60x 8mm 7 inch reels maximum but we will support different line sizes like 12mm, 16mm ....
I have not planned yet how to arrange the different line sizes in the cassettes.
You can fix trays on the workplace.
If the whole feeder had 12mm lines you may get 7 lines in this slot. (16mm 5 lines)
In the starter version only 3 slots are equipped but you can buy additional feeder cassettes for the 3 empty slots.
We will support higher components like capacitors.
The nozzlechanger holds 4 different nozzles to support all sizes of components.
It could place parts down to 0402.
The speed should be like a TM240A my machine places average 1800 components per hour.
We should reach >1000 components per hour if vision system is used.

Re: SMT700 New vision based SMT machine

Reply #6
it sounds like you've made some significant improvements to the TM240A concept while keeping price in check - very nice!

Re: SMT700 New vision based SMT machine from SMALLSMT

Reply #7
Thank you for compliment :)
You can change equipped feeders and the software will support import function for feeder configuration.
Our design is much more stable and more precise. We use 20mm and 16mm guidings instead of 12mm and 10mm at the TM240A.
My TM240A has head vibrations at mid speed movements.

You get a greater value and much more flexibility.
I use SMT machines for about 10 years and tried to integrate all nice features at a reasonable price.
We will release more information about the next steps in the next month's.
         
All suggestions for improvement are welcome.

Re: SMT700 New vision based SMT machine from SMALLSMT

Reply #8
Excellent work, I can't wait until its 100% finished. There are two reasons I haven't bought a TM240, firstly the feeders are not removable so I would have to stack it twice or buy two machines, at least five of my boards have 50+ SMT components in the BOM. Second, lack of vision system - I do not want to have to closely examine each board and line up components that are misplaced; I know that most straighten up on reflow but my experience is that some don't. This machine seems to address both shortcomings.

By way of suggestions I would like to ask for very good support and ease of use with Cadsoft Eagle; this machine is at a price point where many of your customers will be using this program rather than Protel and if you make it easy for them it will be talked about on the forums and should lead to many extra sales - I would be your first one ;-)

Good luck!

Allan

Re: SMT700 New vision based SMT machine from SMALLSMT

Reply #9
Dear Allan,
thanks for the praise you are my target customer.

The file import supports at this time Altium Designer, Protel 99SE, Power_PCB_180, Allegro but it should be no problem to support Eagle files.

I had the same situation like you and bought several machines having different problems.
A professional used machine  having 60 feeders will cost > 25000USD.
But how could you transport the machine in your office because the size is too big and the weight above 350Kg.

So I decided to build a machine that support all needs for office use and doing prototypes to several hundred boards a day.

We could scale up the platform to reach a fast servo driven machine at a price point below 10000USD.
Every machine I found below 10000USD won't meet the European machines directive and many of them had no CE conformity.

If you use the TM240A you need to adjust some parts and add other by hand but you get always trouble using components in plastic strips.
I placed 150000 parts using the neoden machine but it's use is very special.
The best placement result you get if you are using resistors and capacitors having cardboard strips.
The second main problem is the board alignment.
I use a sheet metal that carries and align pcb by using adjustment holes of  the pcb.
Now the placement errors depends on component pickup and the movement under the nozzle if the head moves. (vibration problems need to reduce speed)
To minimize the error you need to adjust the pickup position on each feeder line but you have no laser support doing this.
The SMT700 use the down looking camera to help you aligning the feeder pickup point.

You can place more than 60 different components using pickup points on the SMT700 workplace.
You can define a feeder at every position the machine head could reach to use tray or cut strips.

We invite every Neoden machine user to use this machine as upgrade path.

Please support me giving your excellent comments and suggestions.
Best regards
Michael

Re: SMT700 New vision based SMT machine from SMALLSMT

Reply #10
A detail picture showing head assembly.


Re: SMT700 New vision based SMT machine from SMALLSMT

Reply #11
[quote author="supertronic"]<snip>
A professional used machine  having 60 feeders will cost > 25000USD.

But how could you transport the machine in your office because the size is too big and the weight above 350Kg.

<snip>
[/quote]
Good day Michael,

Used machine prices vary and so perhaps where you are they are over $25K, but this is not always the case.  For example I purchased a excellent used (low hours) commercial PnP (Juki/Zevatech PM575) with top and bottom vision, BOC (board offset correction), laser scanning (parts are 3D scanned for deformity before placement), and with about 50 feeders (all types, vibratory base, shaker tubes, jedec trays, etc) for $10K.  In fact one can find some older Quad machines for $3K-$5K if one is patient and looks hard enough.  When I was looking I found a number of commercial machines in the $8K to $13K price range and most of them came with some level of feeder assortment (...I even found 3 Siemens Chipshooter for $5K each and these canplace 20K parts per hour! ... these units were way too big for me and the effort to use and setup was more than I wanted).

The other issue you mentioned is the size and weight... you are quite correct in that commercial machines are usually large and weigh a fair amount (my PM575 weighs 450 kg).  However, they need to have this "heft", as there is a lot of momentum generated during their placement (mine can place 3600 components / hour and the gantry moves about 2 Metres per second) and so the weight is needed for stabilization.  The other issue could be power.  Some Commercial machines require 3-phase power (i.e. Universal GSM) which can be a problem for some... it was for me.

Your design looks excellent and well thought out, however, I would suggest you be cognizant of the price.  I say this, as in my experience once a price is $5K and up, then a firm's (customer) decision to purchase takes a lot more effort... i.e. more people within the firm need to review/approve the purchase, after sales and long term support costs are evaluated, risks of purchase from a new firm, lack of third party support, etc.  As I said, I do not know the European market nor your target market and so perhaps your price point is a bargain.  Here in Canada/US the market is indeed quite aggressive. 

Anyway, I hope you do not take my comments as being negative.  I simply wanted to provide you with some feedback and my experience.  Best of luck on your project!

Cheers,

Sam

Re: SMT700 New vision based SMT machine from SMALLSMT

Reply #12
Dear Sam,
Quote
mine can place 3600 components / hour and the gantry moves about 2 Metres per second)
your old machine doing 3600 parts in an hour under full vision placement?
Our machine can move 2m per second but this will not help if you move large parts.
We doing a design for a prototyping machine new, high precision and at a reasonable price.
If you buy a 120000USD machine used at 5000USD there is not much in best condition and  you will need to use an old dos or unix interface to program. Having thousands of parameters to adjust.
What's the price for spare parts if the machines stops running?
For 6000USD you can buy 20 feeders for your machine?

Quote
The other issue you mentioned is the size and weight... you are quite correct in that commercial machines are usually large and weigh a fair amount (my PM575 weighs 450 kg). However, they need to have this "heft", as there is a lot of momentum generated during their placement
It depends on how much mass needs to be accelerated!

Your machine could not be used under office conditions to big to heavy and the size is to big to move thru an office door.

Our target customer needs a small pick and place machine easy to use and doing the most prototyping jobs.
The SMT700 machine weight will be near 100Kg.

Quote
risks of purchase from a new firm
I am doing electronics design since 1988 how much experience do you want to see.

Thank your for the suggestions.

Best regards
Michael

Re: SMT700 New vision based SMT machine from SMALLSMT

Reply #13
[quote author="supertronic"]Dear Sam,
Quote
mine can place 3600 components / hour and the gantry moves about 2 Metres per second)
your old machine doing 3600 parts in an hour under full vision placement?
Our machine can move 2m per second but this will not help if you move large parts.
We doing a design for a prototyping machine new, high precision and at a reasonable price.
If you buy a 120000USD machine used at 5000USD there is not much in best condition and  you will need to use an old dos or unix interface to program. Having thousands of parameters to adjust.
What's the price for spare parts if the machines stops running?
For 6000USD you can buy 20 feeders for your machine?

Quote
The other issue you mentioned is the size and weight... you are quite correct in that commercial machines are usually large and weigh a fair amount (my PM575 weighs 450 kg). However, they need to have this "heft", as there is a lot of momentum generated during their placement
It depends on how much mass needs to be accelerated!

Your machine could not be used under office conditions to big to heavy and the size is to big to move thru an office door.

Our target customer needs a small pick and place machine easy to use and doing the most prototyping jobs.
The SMT700 machine weight will be near 100Kg.

Quote
risks of purchase from a new firm
I am doing electronics design since 1988 how much experience do you want to see.

Thank your for the suggestions.

Best regards
Michael[/quote]

Good day Michael,

As I said, please do not take my previous comments personally.  They are not intended as a slight against your product, but were meant to give you some insight as to how other users may think.

The specs are what my machine can do, but I have not really timed it, as I only use it for prototyping (i.e. a few boards here and there). For my needs I really needed the top/bottom vision, etc, as I use BGAs in a number of my designs.

As for my comment about the gantry speed... this is simply a comment and was not meant as a comparison and the reason why the (older) machine(s) needs to be the size/weight that it is.  That is not to say that yours could do the same or better, but simply a reason that the older machines are as big and bulky as they are.

As for the older O/S used by the machine... a non issue for me.  It is certainly not a point and click operation, but again it has not been an issue.  One could argue that the older O/S, etc is unsupported, etc... However, it works well and is backed up and so the risk is relatively small.  If I want, I could simply modernize the system, re-write the operating software, and re-use the mechanical infrastructure as well.  Would I do this?  Probably not, as I have too many other things to do.  Having a new machine with modern software would be nice... there is no argument here. 

As for the spare parts... again... a non issue.  One of the benefits of using an established commercial machine is that there is a good support infrastructure.  I can get parts pretty much anywhere (Canada, US, Overseas) and there have been a number of used non-working machines available that can be purchased cheap and then used for spare parts.  Feeders, etc are also not a problem to get (about $100 or so for used feeders).  I do not know how you intend to distribute or provide after sales support for your product, but it is a concern to some users/firms.  For example are you planning on selling direct or will you be reselling through a dealers, etc.  What about support? Are you planning on offering support simply by you or your firm, or will you have support/repair centres in various locations?

For $10K I bought a used low-hours machine (from the original owner) that has more feeders than I need and virtually looks like new (not one scratch).  The owner took excellent care of the machine and provided training to me when I purchased the unit.  The feeders are separate assemblies and so I can leave standard components on them all the time.  I have 8mm, 16mm- 24mm, and 32mm feeder reels.  I also have an assortment of jedec trays, shaker tubes, and two vibratory base.  The unit can be set up with conveyance so that the picked board can be routed to an oven.  I can load parts on all four sides and I think the machine number of reels I can load is over 100.  This depends upon how the machine is configured (i.e. if shaker tubes are installed, etc).  The point is that one can find a commercial machine that has a lot of stuff for far less than the $25K you originally stated.

You are absolutely correct that the commercial machines are bigger and heavier and cannot be used in locations or users that have limited space.  In these cases a smaller machine like yours would be perfect.

I recognize that you have been designing since 1988, but overall some firms look at other aspects.  For example, some companies consider smaller firms to be a risk.  For example the smaller firm could close or go out of business for financial or health reasons, loss of key staff, litigation, etc.  I do not know the size of your firm, but I am small firm and these are issues I face daily despite being in business (Engineering) for 20 years.  What I have found, and presented to you, is that if the price is over $5K or so, then the product falls under greater scrutiny and usually requires more than one or two people to make a decision. When this happens the "risks" I have stated come into question.  This may completely unfair and unfounded, but this is what I have found.

Again, please do not have my comments deter you on your project... the comments are simply feedback and/or a perspective that you may or may not have considered.

Cheers,

Sam

Re: SMT700 New vision based SMT machine from SMALLSMT

Reply #14
Why did you design it with only a single head? Wouldn't having 2 heads and the advancing pin on the back of the gantry, like the TM240A, result in a much higher component placement speed?