Charmhigh T560P4 PnP Problems

[Ribster]

I've had the machine for 1+ years now. And i can say that my feeders do not have this issue. They were also supplied from charmhigh.
It's a cheap chinese machine. So expectations should be handled accordingly.
I've had 1 pick head smash into my pcb because i did not set the pick height of a 7mm smd cap correctly. This mangled the hollow stepper shaft.
I've ordered 2 new ones and replaced it. No issue there.

I do have 1 feeder bank position that does not work, this is after 9 months of use. I assume that the valve / coil failed. Have not yet replaced that part.

The pressure reduction he put on the individual feeders is a fix he's using for the pressure. If you just decrease the flow of the air to the feeder, it will just reach the pressure more slowly, as he said.
Now, he can take the edge off with this reduction, but will need a non-return valve (don't know what it's called in english) in parallel over the reduction. So he only limits one direction.
That said, altering the individual feeder pneumatics is not the way to go.

On my feeders, the pressure at .6 MPa is just perfect. I've had some issues with the 24mm/16mm (can't remember) feeders in combination with a plastic usb-c connector.
In this case the plastic tape falls down on one side in the feeder. But hey, it's not a big deal.

Doing 0402 parts, which i almost always do, is pretty stable. Sometimes i have a resistor on it's side, but in the correct rotational direction.
On the most part, can't complain for the machine. It does what it needs to do for my needs.

IMHO, there is a lot of energy going on with the youtube making. Their experience is valuable. But it's a sample size of 1.
There are others out there that are using the machine successfully.
It's certainly no fun if you get the machine and there are issues, but, it's expected. Try and be creative, get to know the machine and do some manual pick/placing when required.

[SMTech]

Our relationships with companies are not forged when tings are going well, rather by how they respond when things go poorly. We recently had two issues with PCs the first was a HP unit that stopped POSTing after a windows update, in 2 days HP had sent someone out with a new motherboard (3 year onsite warranty option taken up by the equipment manufacturer that put this PC in their kit) . The second was a new laptop delivered damaged from Dell, 3 weeks on, we are still waiting for a refund & for them to arrange collection, you can guess who won't get another sale. The impression I get with all of these machines is that on the whole if you have a small issue and ask for a part, you seem to get a response and a spare part. This is as you'd expect and what is required.

What has happened here is the machine has much more than a minor fault and their response has been poor to say the least. As there are no proper on-board diagnostic tools on board and some of the issues could be caused by different problems, China has no way of helping remotely, they are are relying on guesswork tainted by language barriers and client ideas, and somewhere in the conversation someones fur got rubbed the wrong way and there doesn't appear to be a way of getting back to an even footing. I've watched a few of his videos, you can hear just how upset and angry he is even when he's trying to laugh it off its quite hard when you feel that way to deal nicely with the people causing the issue, a 3rd party maybe, but the lack of one of those is just one of the things that makes this machine cheap.

[Mark19960]

Been watching this situation from afar and I know a few people with these Charmhigh machines and the various clones.. or is the Charmhigh a clone? who knows.
With any larger scale projects I just get the boardhouse to assemble it.
They will do it cheaper and faster than I can.

My issue is that Charmhigh won't get the thing sorted.
And if they can't or won't get the thing sorted you can't rely on them for support period.
Not the first time I have heard about them not supporting or poorly supporting the end user and support is kind of important with these machines.

That said, for my proptyping I pulled the trigger on another machine the other day and I had been considering one of their desktop machines.
Charmhigh lost out on that one because I don't think I could rely on them to support the machine.
Not going to have that nightmare and be out 5k nicker!!!!

[SMTech]

Been watching this situation from afar and I know a few people with these Charmhigh machines and the various clones.. or is the Charmhigh a clone? who knows.
With any larger scale projects I just get the boardhouse to assemble it.
They will do it cheaper and faster than I can.

My issue is that Charmhigh won't get the thing sorted.
And if they can't or won't get the thing sorted you can't rely on them for support period.
Not the first time I have heard about them not supporting or poorly supporting the end user and support is kind of important with these machines.

That said, for my proptyping I pulled the trigger on another machine the other day and I had been considering one of their desktop machines.
Charmhigh lost out on that one because I don't think I could rely on them to support the machine.
Not going to have that nightmare and be out 5k nicker!!!!

Indeed but how many of the others would act in the way we would like with an issue of similar scale? The machine in question clearly has an entire batch of crap feeders, which has to mean other people do too but also has dodgy drives/wonky frame/poor belt tension affecting the overall accuracy.

I can only reiterate what I've heard about a certain other brand, the reseller in the UK was so unhappy with the build quality and support they received they ended up refunding multiple installs out of their own pocket and took them off their books. In that instance there were two level of support open to them, China and the European distributor who was somehow still in the mix and had a very european shrug response to pretty much everything.  On that basis I don't think you can RELY on any of them, they are just machines that will do a job and you need to be able to handle significant self diagnosis & solving of issues and waits for parts if issues crop up.

I can rely on our Essemtec, its a global company with spares in multiple locations, dedicated support staff, knowledgeable trained distribution channels and everything on the machine is configured for remote diagnostics. All of that we have to pay for, its comparatively expensive but by no means the most expensive option and critical to the operation of the business, the cheaper option is to buy a 2nd user machine that also has all those support avenues still open. If this Essemtec machine had arrived working that poorly, I wouldn't have just had the local tech here, there would almost certainly be an engineer from Switzerland on-site too.

[nightmar3]

Hi Everyone,
Just to let you know. I bought the same machine about a couple months ago and finally got around to getting it up and running unfortunately....I am having the same issue as Seon with the part bounce. Not only that but some of my feeders are completely useless. I have found about 5 of my 3 (8mm feeders) are not able to keep the air and I believe it is because of a faulty solenoid. I will keep you guys posted since Seon and Zero have not provided any update. I pinged them with a couple of updates the past couple of days I will see if I get different feedback than Seon

[Whales]

He's now claiming he will take all of his videos down if Charmhigh sends him working electric feeders.  Sounds like a dangerous promise, but given his claims that Charmhigh won't talk to him it might not be much of a risk.

[Whales]

New claims: machine mysteriously picks up and places parts from feeders that are not even being used for the given board.  Vacuum (loss?) detect doesn't work.  Vision is unreliable.  Machine behaves slightly better after reboots.

I was thinking the head accel/decel might be too high and parts are flinging off, but that doesn't explain why his 0402 resistors started reliably appearing in one spot.

 

[Whales]

I really want to see footage of this machine failing to place parts.  Rhetorical questions:

 - Can you record/tap off video from the vision camera whilst it's running?
 - Can you tell the machine to run the head a lot slower?

[Whales]

Hi Everyone,
Just to let you know. I bought the same machine about a couple months ago and finally got around to getting it up and running unfortunately....I am having the same issue as Seon with the part bounce. Not only that but some of my feeders are completely useless. I have found about 5 of my 3 (8mm feeders) are not able to keep the air and I believe it is because of a faulty solenoid. I will keep you guys posted since Seon and Zero have not provided any update. I pinged them with a couple of updates the past couple of days I will see if I get different feedback than Seon

Welcome to the forum nightmar3.

Faulty air solenoid?  Hmm.  I remember from repairing printers that simple/tiny things can throw solenoid timing off, eg steel-on-steel contact where there used to be a small foam or paper spacer (residual magnetism in the touching steel parts makes them stick for a little longer).  It would be worth checking if they are all the exact same solenoid design or not.

I hope things go well for you.  Keep things polite and best of luck.

[Whales]

We're back with "NEWS! Delivery from Charmhigh!"

Summary:
 - 3rd party now helping mediate between Charmhigh & Unexpected Maker
 - UM has received replacement feeders (electric instead of pnuematic) from CH
 - One new feeder tested appears to be much better (not bouncing/losing parts), but UM is still to try them all
 - UM says CH won't acknowledge any problems with the machine itself, only the feeders, so he is only negotiating about feeders
 - UM says he will take down all videos + forum posts critical of CH if this fixes the major problems with the PnP.
 - UM is really pissed and wants to "move on", not keep making videos about CH, doesn't want his channel to become centred around CH critique



I hope this fixes his problems and he doesn't end up walled by new ones.  It's risky to say the problems are only in the feeders, but I'm not sure if the other side will appreciate this risk. I hope both sides can work it out.

[seon]

Well, here's the finale! It's a wrap... seriously, I've tapped out!



Seon
Unexpected Maker

[Mecanix]

Well, here's the finale! It's a wrap... seriously, I've tapped out!

Seon
Unexpected Maker

Nice wrap-up and lucky you to have something that can be worked out, for that sort of price range anyway. Still on the All Mighty Tweesers here so I'm kinda jealous I gather from a quick glance at your thread you had a bit of bumpy ride to get it to work, watched that last vid and would like to suggest ideas on the subject of jumping parts and misalignment;

- "Jumping Parts". Add weight. Enclosure construction appears weak & light sheet metal (no damping and prone to nasty harmonics). Without over-killing it, add a great amount of weight to it while carefully making sure you aren't bending its axis. Have e.g. a precision level on its axis, monitor that and gradually add that weight in.

- "Homing & Repeatability". I'd be surprised the engineering dept have chosen thermally stable materials to build this machine, let alone adding temp sensors and software compensation/coefficients required to cope with this like other machine-tools have. That said, always "use and calibrate" that machine at the same time of the day or when its within a 2~3dC temp difference from the previous cal/run only. It won't 'magically' fix it however improve greatly on the repeatability side of things

- "Accuracy". Google the term Axis Backlash and learn how to measure it. You'll need a dial-indicator (few dollars) with a magnetic base ideally. Process is as simple as moving the axis in one direction for a known amount of distance, then back to original starting location (note the offset on the dial indication, that's your backlash). Once you know where and how much, you can compensate via software whether manually before use or automated (not sure they have this feature, should have).

Have plenty of other ideas for you but that would be a good starting point. Hope this helps a bit. Have fun.

[ar__systems]

I've had similar problems with CL feeders on my TVM920. Some appears to be worse than others. I too experimented with flow control and pressure. It can be made to work with asymmetrical valves (restricted flow out, unrestricted in). In the end I figured that rubber band around the feeder (to push the top moving plate down) is easier. Rubber bands keep breaking but I don't care. That trick virtually eliminated the problem for me. 

Few suggested weight of the machine - no. This has nothing to do with machine. It is a feeder issue.

[kylehunter]

I've had similar problems with CL feeders on my TVM920. Some appears to be worse than others. I too experimented with flow control and pressure. It can be made to work with asymmetrical valves (restricted flow out, unrestricted in). In the end I figured that rubber band around the feeder (to push the top moving plate down) is easier. Rubber bands keep breaking but I don't care. That trick virtually eliminated the problem for me. 

Few suggested weight of the machine - no. This has nothing to do with machine. It is a feeder issue.

Care to show some pictures of the rubber band method? We have 0 issues with parts jumping on 0603s and up, but on 0402s it can be pretty bad at times.

[SMTech]

Well, here's the finale! It's a wrap... seriously, I've tapped out!

Seon
Unexpected Maker

Nice wrap-up and lucky you to have something that can be worked out, for that sort of price range anyway. Still on the All Mighty Tweesers here so I'm kinda jealous I gather from a quick glance at your thread you had a bit of bumpy ride to get it to work, watched that last vid and would like to suggest ideas on the subject of jumping parts and misalignment;

- "Jumping Parts". Add weight. Enclosure construction appears weak & light sheet metal (no damping and prone to nasty harmonics). Without over-killing it, add a great amount of weight to it while carefully making sure you aren't bending its axis. Have e.g. a precision level on its axis, monitor that and gradually add that weight in.

- "Homing & Repeatability". I'd be surprised the engineering dept have chosen thermally stable materials to build this machine, let alone adding temp sensors and software compensation/coefficients required to cope with this like other machine-tools have. That said, always "use and calibrate" that machine at the same time of the day or when its within a 2~3dC temp difference from the previous cal/run only. It won't 'magically' fix it however improve greatly on the repeatability side of things

- "Accuracy". Google the term Axis Backlash and learn how to measure it. You'll need a dial-indicator (few dollars) with a magnetic base ideally. Process is as simple as moving the axis in one direction for a known amount of distance, then back to original starting location (note the offset on the dial indication, that's your backlash). Once you know where and how much, you can compensate via software whether manually before use or automated (not sure they have this feature, should have).

Have plenty of other ideas for you but that would be a good starting point. Hope this helps a bit. Have fun.

Its a closed system, he can look for loose belts & dodgy couplers, and that is absolutely the clients job when you buy a cut price thing from China, but tweaks involving firmware/software & drive control are a dead end and a waste of time and money to chase down. I think you'd need to take this thing to bits and reassemble it to even get close to having a full picture of how many things are adding up to make it this bad.

[Mecanix]

Its a closed system, he can look for loose belts & dodgy couplers, and that is absolutely the clients job when you buy a cut price thing from China, but tweaks involving firmware/software & drive control are a dead end and a waste of time and money to chase down. I think you'd need to take this thing to bits and reassemble it to even get close to having a full picture of how many things are adding up to make it this bad.

Sure is. OP seems shattered, was just trying to chime in some. I personally haven't seen one of those machine in person but read from someone who've just had his $14.5k gone to the bin, literally. That thing came with cheap rubber timing belts without inner cords, stepper motors (I'm serious lol), no encoders, an offset in the linear rails squareness, zero motion profiling in terms of acceleration & deceleration, etc etc the non-sense list never ended. Funniest; had acetal (POM) made parts all over it i.e. a material with a thermal expansion coefficient greater than any others. Could pick-and-place a SOP-8 but positioning/repeatability was based "on luck" all depending on the ambient temp on that day, forget about 0805 and below (8 score out of ten, kinda set-up...).

I could be dead wrong but looks like there some sort of scam going on. Appears that a few figured out "Makers" are on the rise and the need for P&P machines are also, taking advantage of that "trend" so to speak. Not sure if its because I've been looking into gearing up myself but I seem to find those Chinese look-alike PnP machine everywhere I look. And they aren't exactly cheap considering the good ones they sell cost $500~600 in parts, optics and sheet metal (max), add $100 for that random worker who assemble it :/ *Beware folks lol*