[MrMetric]

Hi.
I'm on the hunt for a wireless Centroid pendant. And, yes, even though it is only a $120 XHC PHB04B pendant, you have to use *their* pendant. :-(. Does anyone here have one that they want to sell? It doesn't have to be a hanger queen. PM me if you do, please. If I find one, I'll post here that I'm no longer in need of one.

Thanks!

[allencncrouter]

check this Contact with me on whatsapp +8618560201508

[MrMetric]

Unfortunately, that will *not* work on the Centroid system. There would be a whole host of different pendants I could use if I the interface were I/O ports, but the Centroid is all USB based. That means a DLL is needed to interface with the pendant. This is how the XHC PHB04B works, and it allows Centroid to "lock" pendant support. Unless you use their product, you really cannot do much else.... There *might* be some wiggle room by using the encoder port, but I don't think it will be all that easy. For instance, I don't think you'll be able to update the display on your device with the position information, etc.

[joeavaerage]

Hi,
either buy a Centroid pendant and be done with it.........or make a manual one.

I use Mach and have done for years. I used a VistaCNC P1A pendant for many years, and it too was USB connected. VistaCNC wrote a plugin that allowed it
to communicate with Mach.

After about eight years it started going flaky, I had basically worn it out. I should really have just bought another, they are in effect cheaper now than when I first
bought it......but I decided to make my own. I got a really nice Japanese made MPG wheel and built my own.

I have deliberately kept it pretty basic in order to avoid using to many wires (equivalently inputs of my controller) . In its current form it has an axis switch for X,Y,Z,A,B and C axes,
the MPG outputs and a spring loaded switch to select Jog Velocity mode. It requires two wires for the MPG, three wires for the BCD coded axis select, and another two wires for Velocity/Step modes, ie
seven signal wires (seven inputs) and another two wires for the power supply and supply return.

It works well, I've been using it for about two years now. The MPG wheel cost $60USD (second hand), so I guess the whole thing cost about $100USD, not counting my time.

Craig

[MrMetric]

Hi Craig,
Thank you for the response and for sharing your experiences.... I have a bit of a distrust of "closed" systems, so I'm not thrilled about Centroid's decision to lock out pendants. However, their system does generally seem to get pretty decent reviews. My biggest concern about "rolling my own" is more whether it is really feasible given Centroid's proclivity towards protecting their turf. Not having a display that shows location is, frankly, not that important. The #1 use case for my having a pendant is generic JOG moving and touch off creeping when finding the part zero. Neither of those really requires a display showing the location. And MPG is critical, though.

The Centroid Acorn is, unfortunately, very limited in its Inputs and Outputs. To that end, the USB based pendants have an advantage in that they don't eat into those meager I/O ports. However, I do have my concerns about having an interpretive button press. That seems a little risky, but it isn't any riskier, frankly, than the touch screen on the Centroid because at the end of the day that is also USB based.

Anyhow, thank you for the thoughts. I'd be happy with a 60 to 100 dollar pendant. It is the 360 dollar one that costs as much as the controller itself which has me coughing.

[joeavaerage]

Hi,

I have a bit of a distrust of "closed" systems, so I'm not thrilled about Centroid's decision to lock out pendants

Have Centroid 'excluded' other pendants?. Mach is a commercial software product and they might have chosen to limit the access to their code so that manufacturers could
not make products using Mach....but they chose to do otherwise. They share their code with manufacturers so that the CAN manufacture supporting controllers pendants and whatever.

Have you asked Centroid about about getting sufficient info about their software so that you could write a plugin? They are at liberty to decline of course, and indeed they may, but before asserting that they have
'locked' anyone out it would be polite to ask. Second question is why did you buy Centroid in the first place if you were concerned that they prevented competition in the hardware market?
One of Centroids strengths is that they are a one-stop-shop, ie software and hardware. This ensures the best possible compatibility and least hick-ups for users. To then complain that you cannot use
competitive hardware would dilute that strength.

Not having a display that shows location is, frankly, not that important. The #1 use case for my having a pendant is generic JOG moving and touch off creeping when finding the part zero. Neither of those really requires a display showing the location. And MPG is critical, though.

My P1A pendant had features that I never used or hardly ever used. Things like Jog Increment, Mach Screen Select and all sorts of other stuff. After eight years of use I determined that I need:
1) Be able to select each axis
2) Have a silky smooth MPG
3) Have a Velocity Jog/Step Jog mode switch

that's it!!! I do not need a screen, I'm looking at the bloody workpiece, not some ridiculous screen that only ever replicates whats on my PC anyway. I don't even have an Estop on my pendant.
I do have a big red EStop button on my machine so I don't see the need for another on my pendant....I never used the P1A Estop, so why would I waste my time putting one
on my pendant? If you strip the pendant down to what you actually use and need then it becomes vastly simpler.

If you follow my logic then you can reduce the number of wires (equivalently inputs) dramatically.

If you have a three axis machine then two wires (BCD coded) are all that is required to select an axis. Two wires (A & B) are all that is required for the MPG. I don't know how you select
Velocity/Step jog in Centroid or even if they have them......so for simplicities sake don't bother with them.
Thus you would need four inputs....does the Centroid not have four spare inputs?

Craig

[MrMetric]

Yes, for the Acorn systems, they apparently (95% sure, but I am *not* an expert on them yet) have precluded the ability to integrate with their product. Mach, I believe, is much more 'open'. In the discussions I have had, they do state that a USB based pendant can only be purchased from them. Integrating via the I/O lines might be possible, but still needs to be confirmed. The main issue with that is, as mentioned, there are not that many inputs on the Centroid Acorn (8, as I recall). Those are taken up with limit switches and the like.

"Why did I buy Centroid"... Well, that is a very good question. I was originally thinking of a Chinese DDCS Expert, but not for the reasons you might think... My project needs to be completed pretty quickly. Building HID panels is not really feasible. The final product also has to be easily used because it is for my brother, not me. He isn't tech savvy. In fact, to that end, Centroid pushed over the top because they have a conversational system that I think would work well for him. Plus, being US based, I think it has a better chance of having good quality and support. But, going back to the DDCS, the reason why that was considered in the first place is that all the buttons are on the unit already. For the Centroid, it is dominantly touch screen based, which fit the "quick to fit" requirement well. I did not go the Mach route because I haven't been super thrilled with Windows being the main controller (Centroid Acorn uses windows, but the motion side is in a BeagleBone).

What I did NOT realize before I purchased the Centroid Acorn was that I was locked into their pendant. Given that the use of a pendant over DOUBLES the cost of the system (because you need the Pro version of the software too), this little oversight has been a stumbling block. I still think that Centroid has an excellent system, and I don't even begrudge them the need for the Pro system. It is really the pendant price that bothers me, I think.

But that gets to the final point here.... What is really needed? I actually agree with you wholeheartedly. At the end of the day, jogging and inching on the axis is really all I need. Unfortunately, to do this, I pretty much *HAVE* to have a USB based pendant.... USB doesn't eat into that meager number of inputs. So, the display has never been a big interest of mine.

I think that at this point, the biggest limitation to Centroid, and why I probably would not recommend it to someone in the future, is the lack of IO either directly or indirectly (through the use of a DLL, etc). Is this the right system for my brother? Actually, I'm still *fairly* sure it still is, but it will cost him more than I told him it would, and that doesn't make me feel very good. Would it be the right system for me when I convert a machine? Probably not. I have done LinuxCNC before and I fully appreciate the wide open landscape it affords. But I'm a technogeek. I program for a living and I used to design hardware for a living too. I definitely don't mind getting my hands dirty in code or getting much deeper into esoteric configuration stuff than many might be willing to do.

Anyhow... good discussion, Joe. I appreciate your thoughts.

[joeavaerage]

Hi,
yes, it has been a good discussion. It seems I have found someone whom agrees with me that a pendant can actually be very simple, it does not need all the bells and whistles. It seems many a hobbyist CNCer
believes all the bells and whistles are required and desirable, and that wireless is required??? My pendant sits on the bench and I hardly ever move it, let alone exceed its 3m cable. When I use it I am concentrating
on the workpiece and that means ipso-facto that I'm standing right in front of it....wireless not required. It makes me wonder sometimes whether any of these guys do any actual machining. I'm all for anything
that makes making parts easier, but wasting time, money and energy on junk that does not improve my parts....no thanks.

I still think that Centroid has an excellent system, and I don't even begrudge them the need for the Pro system. It is really the pendant price that bothers me, I think.

Is this the right system for my brother? Actually, I'm still *fairly* sure it still is, but it will cost him more than I told him it would, and that doesn't make me feel very good.

I think that your estimate is fair, Centroid do have a fairly complete system ready to go. It does what it claims with little fuss....and OK that does incur some 'flexibility penalties' but does stiil perform
well overall, especially for someone whom it interested in machining rather than techno-babble.

I found myself defending my choice of Mach4, an Ethernet Smoothstepper and my own build breakout board a while back. The combination set me back about $600, and that ascribes a value of my breakout
board of $200. I've been using Mach4 for 8.5years, and daily for the last 3 tears in pursuit of my business, ie $70/year. Just my five servos alone cost $2500, and my new machine (now including my home built
trunnion/fifth axis) cost about $25000. I pay about $500/year for Fusion Basic and then another $1300/year for Fusion Machining Extensions....the point being that my controller and matching CNC software
is the smallest part of the cost....so don't sweat it.

Even if Centroid cost double or triple what you estimated over ten years its still nothing by comparison to the costs of the machine. What matters is not the cost, but rather the usability, and Centroid Acorn enjoys
a good reputation in that regard. I'd say use it, even their proprietary pendant, and the cost be damned.

Craig

[MrMetric]

The irony here is that in this particular case, the machine is less expensive than the Acorn setup, LOL. But that is a bit of an anomaly. Usually it is the other way around....

FWIW, I would actually *prefer* a hardwired pendant over a wireless one. I famously hate AA and AAA batteries because they are always dead when you need them, and when you replace them, they are always leaking crap all over the place! The 1.25V Alkaline battery deserves to die a miserable death. That fate is decades overdue! Now, the 9V, by contrast, actually works well and doesn't generally leak.... In addition, I just feel more confident with a basic hardwired unit. Unfortunately, this doesn't work well in the Acorn paradigm, as I mentioned. There are just too few IO pins on the board and no way to increase the count (another thing that I wish Centroid would adjust, but I suspect it would cannibalize the sales of more expensive units).

Anyhow, I absolutely hear you on the cost amortized over years of ownership. But... there is more at play here than just "me." I would go ahead and say, "yeah, whatever.... I want *some* pendant so just buy it...." But the funds are my brothers and I try to guard those. I'm already funding many parts of the conversion, to be honest. I'm providing the computer, discounting the motor acquisition for him, doing all of the labor/configuration for free, etc. I have my limits to what I am willing to do! Adding another $360 on top of what I'm already giving him on just this machine is more than I really want to do right now. I have my own obligations and needs to consider as well, after all. But, I also know he is spending a decent amount on the whole thing.... It is that type of personal conundrum that is, well, very personal! LOL

Nice pics, by the way.... I always enjoy seeing what people are doing, so thanks for sharing.... I do hope your significant other didn't beat you over the head, however, when she realized what happened to the cutting board!

[joeavaerage]

Hi,

I do hope your significant other didn't beat you over the head, however, when she realized what happened to the cutting board!

When I say I've 'breadboarded' my design, I bloody well mean it!! I have bought half a dozen of them on which are mounted servo drives, power components,the breakout board such as you've seen.
They are really just a cheap piece of plastic, and all the same size. When they are all populated they are slid into a rack. Thus if I want to add something new to the breakout board, just pull that 'card' and there
it is. Or if I want to tweak the servo drive firmware I pull the servo drive 'card'. Just convenient.

Craig

[MrMetric]

I like it.... a true "breadboard", even if you are using mixing the electronic and baking terms (hey! get it... "mixing" is also a baking term). OK.... I digress...

I have an important learning that may help me significantly. I *think* you might be able to add an ethernet based expansion board such as the Ether16. If that is the case, it substantially changes the rather limited I/O that I have been concerned about. That also alleviates some of the restrictions regarding pendants and such. It also makes a lot of sense. Frankly, Centroid would have been crippling the Acorn with an unrealistic number of I/O, IMHO.

Anyhow, I definitely need to learn more about the abilities and capabilities of the Centroid Acorn. It is a work in progress, but if I can successfully install a pendant such as what you've been talking about, I'm a happy camper once again. Axis selection, jogging +/-, multiplier and MPG are really all that I want. And if this is all hardwired, then I am completely fine with that (in fact, I prefer it). These capabilities will take me from "I'm not really sure I want to use an Acorn" to "yep, I think this will work out well."

[MrMetric]

I like it.... a true "breadboard", even if you are using mixing the electronic and baking terms (hey! get it... "mixing" is also a baking term). OK.... I digress...

I have an important learning that may help me significantly. I *think* you might be able to add an ethernet based expansion board such as the Ether16. If that is the case, it substantially changes the rather limited I/O that I have been concerned about. That also alleviates some of the restrictions regarding pendants and such. It also makes a lot of sense. Frankly, Centroid would have been crippling the Acorn with an unrealistic number of I/O, IMHO.

Anyhow, I definitely need to learn more about the abilities and capabilities of the Centroid Acorn. It is a work in progress, but if I can successfully install a pendant such as what you've been talking about, I'm a happy camper once again. Axis selection, jogging +/-, multiplier and MPG are really all that I want. And if this is all hardwired, then I am completely fine with that (in fact, I prefer it). These capabilities will take me from "I'm not really sure I want to use an Acorn" to "yep, I think this will work out well."

UPDATE:
Sadly... no.... The Ether16 is actually the Ether1616, which is another Centroid product. The price of this is the same as the pendant, so there really is no cost reduction here, only a lot more work. My strong suspicion is that Centroid is protectionist on the interface for the Ether1616 in the same fashion as the pendant, so I haven't really gotten anywhere. Darn.... I was hoping that I had found a viable solution. I have to have a heart to heart with the financial power to see which way to go, but I suspect that there is a very good chance I'll be returning the Centroid unopened (if possible). It feels like it is just getting too expensive for this particular machine. That might not be true for others, but here I think it might be.... This is also not a reflection on Centroid. I think it is probably a decent product, although I do not agree with some of their design decisions. People who are doing retrofits want flexibility and options, and Centroid feels like it is a little lacking there. The Acorn is billed as being low cost, and this is largely so if you are doing a very very simple conversion. But even something as simple as a pendant will suddenly get you up to $900. A pendant is a pretty basic add for any conversion, especially when the only other viable option is a touch screen interface. For smaller conversions where you are looking for an easy turnkey solution, I think the DDCS Expert is probably a better way to go. It will cost you roughly half the cost with a pendant and a lot more I/O. Oh well.... there is a learning experience here too. Knowledge is king. And, hopefully this thread will give others ideas to think about when considering their direction. Centroid *is* a decent solution, but I don't think it fits all scenarios.

[joeavaerage]

Hi,
I am sort of surprised that you would consider ditching the Acorn on the basis of cost. You say $900 for a solution at includes a pendant, and I think to myself ......yes that sounds about right.
Caveat: I know nothing about Centroid Acorn other than that I have read both on the forum and Centroids marketing material.

I am much more familiar with Mach4. Mach4Hobby has a perpetual license (for up to five machines) of $200USSD. I use an Ethernet Smoothstepper, $225USD. If you were to use an MB3 breakout board ($180USD)
or make your own as I have done, I will assume a value of $200USD. I made my own pendant, and really the cost of it is dominated by the MPG. If you want a really slick MPG like a Fanuc wheel be prepared to pay
hundreds. Having said that perfectly useful Chinese made wheels are $20-$50. I ascribe a value to a simple homemade wired pendant at $100USD.

Thus the total for the CNC software, the motion controller, breakout board and pendant is $725USD.

The ESS has 51 inputs and outputs. There is a limited amount of flexibility as to the balance of input verses outputs. In my case I have favored inputs and have maximized those, thus I have 31 inputs
and 20 outputs. I have a six axis breakout board so 12 of the outputs are just the Step/Dir for those axes, so you cannot afford to be cavalier with IO, but that does give me 31 inputs.
To date I have nine inputs for Home and Limits of the three linear axes, a probe, an Estop, and a further seven inputs for my wired pendant and six alarm signals (one from each servo) which leaves me
with seven inputs to spare. I would say that the ESS is adequate for IO, not a surfeit, which would be nice, but adequate.

Mach4 will run on any old PC Windows 7 or better. One of the kids cast off school laptops will do the trick. My machine runs on a dual core Atom single board PC with 4G Ram, no graphics card,
and a 64G SSD. It hasn't got the power to 'pull the skin off a rice pudding'....and I've been using it for ten years. The only time I notice or rue its lack of power is when I load a largish
toolpath, say 5M or more....it loads slowly, up to a minute or so. Thereafter it runs fine. The point is that you don't need much of a PC to run it. I suppose I could run a touch screen,
but the standard Mach4 screens are fairly tightly packed, and so when I used my i7 laptop (on occasion) I found the touch screen to be less than useful....you'd hit a button only to trigger the neighboring
feature!!! Of course Mach4 has a complete screen editing suite so you can edit or even devise your own screen set at will....but for all I found stabbing your oily fingers all over the screen was unsatisfactory.
So I do not have a touch screen and have little inclination to get one. A proper industrial track ball would be nice, but I can always find more important things to spend my money on than something that
is merely 'nice'.

Mach4 has a very useful and versatile scripting language with a fairly complete API, and as I've already mentioned a good screen editor. All-in-all it is flexible and customizable solution.
LinuxCNC still holds the record for flexibility, but that flexibility comes at the cost of coding complexity.

UCCNC is a solution you might look at. The cost of the software (perpetual, per machine) is $60USD. You are required to use one of their motion controllers, but they sell very useful motion
controllers for under $200USD, some of which require a breakout board, but at least one model does not. UCCNC is competitive with Mach4 in terms of features. I suspect Mach4 has the edge
in scripting and screen editing, but those really appeal to technogeeks, so that advantage is moot for regular CNCers. All-in-all a very useful solution at a cost saving over Mach4 or Centroid Acorn.

LinuxCNC is free, and a useful motion board is around $250USD. On the basis of cost it outperforms Mach4, UCCNC and Centroid Acorn. Anyone whom is familiar with Linux and/or coding
in C linuxCNC is a very good choice but is not as popular with non-technogeeks.

Maybe you should have a look at some of the Gcode sender types, GRBL is probably the most mature of that type. I am fairly dismissive of Gcode senders and have described then as a toy, but for
all my prejudice they are effective and cheap.

Craig

[samco]

LinuxCNC is free, and a useful motion board is around $250USD. On the basis of cost it outperforms Mach4, UCCNC and Centroid Acorn. Anyone whom is familiar with Linux and/or coding
in C linuxCNC is a very good choice but is not as popular with non-technogeeks.


Craig

Just to be clear.. Between hal and classicladder - 90 to 99% of what most people need to do with linuxcnc don't need to program in C..

sam

[MrMetric]

It is actually closer to 1K, but I understand your point. However, I'm looking at the composite cost. Another $250 for closed loop steppers, $125 for a monitor, $200 for cabinets and other associated work, and now I'm up to about $1600 for the electric package to put onto a machine that cost $800. That is the stumbling point I have. And, at the end of the day, I'm limited to 8 inputs and 8 outputs with the Centroid which is pretty weak, IMHO.

So, yes... I'm coming to the rapid conclusion that Centroid Acorn is not nearly as flexible at the end of the day. And neither is it as inexpensive as it seems to be popularized as. Now, that price *would* give me the Centroid pendant. In my mind, that is kind of mandatory for the exact reason that you have stated. I really don't trust the idea of "hitting the right spot on the screen" versus a mechanical switch. I feel the same way about touch panels that have zero tactile feedback. Heck, I'm not even a fan of any membrane switch.... There is a reason why the laser drawn keyboards have never taken off. People need to feel the position on a keyboard to be able to type effectively. The same applies to membranes.

I can handle LinuxCNC with no problems. The configuration side, code side, etc. I don't typically code in C or C++ very much anymore, but I do code on a daily basis (in fact, I'm taking a 'break' from doing just that while I write this). However, this machine is not for me... This machine is for my brother and he is an utter technophobe. I cannot even imagine what a nightmare it would be to support him if I handed him a LinuxCNC machine!