What's new
What's new

Brother Speedio 4th Axis Simultaneous Programming

StirlingMachine

Aluminum
Joined
Apr 19, 2015
Hey Folks, got a Speedio S700X1 with a Brother TL200 rotary last month, and trying to figure out how to get it to handle 4th axis simultaneous work properly. Basically what I've come to after a day of digging is that it only supports feedrate input for the 4th axis in degrees per minute. I'm running HSMworks/Fusion, and the off the shelf speedio post doesn't have much to offer there. It just spits the programmed IPM feedrate and then the rotary runs super slow. Anyone have a functioning HSM/Fusion post that supports 4th axis simultaneous work they'd be willing to share?

I'm digging through the Post editing manual now, and it's looking like there may be some stuff I can pull from a Haas 4th axis post that will convert IPM to DPM feeds, but it'd be a whole lot simpler if there was one already out there :)

Cheers all,

Devin
 
that seems hard to believe. I've been considering a new Brother, but if that's true, could be a deal breaker.

It does seem quite odd, I can only assume it is one of the quirks of Brother Industries extreme caution in regards to ITAR regulated technology export.

Having said that, we know the Speedio is perfectly capable as a full simultaneous 4th machine as it’s a common use case, it’s just a matter of getting the Autodesk post dialed in. Lack of inverse time feed is a head scratching quirky annoyance, but not a deal breaker.
 
Thought I'd update this with progress for anyone who comes across this later. As of now Speedio will only accept simultaneous rotary motion in degrees per minute. That in itself isn't too limiting so long as your post will output it. Inverse time would be more accurate from what I've learned (I'm new to rotary work), but this does work.

Buuuuuut, here's the rub. If you run the speedio in Inch Mode the maximum programmable feedrate with Mode B2 installed is 1980IPM. Even though the rotary is reading DPM the control still alarms on any call with an F over 1980. So you can do the math but 1980 DPM is equivalent to around 5.5rpm. It's not that fast. It is usable, but not ideal.

The workaround. If you run in Metric Mode the same rules apply, but just due to the nature of millimeters being very small, the max programmable feedrates are much higher. On this machine as I have it set up 50,000 mm/min is the max feedrate. My rotary runs at 100rpm max, which calculates out to 36,000 degrees per minute. So running in metric you could cut at full 100rpm speeds simultaneous. Switching over to metric is a small hassle, but a hassle nonetheless.

I do hear rumors the support for inverse time feedrate support is coming, or at least being requested of the boys upstairs. I don't know the validity of that though.

That's what I have at the moment.
 
I very much doubt Brother will switch to inverse time from degrees per minute. The majority of the world works in metric and 100rpm in a cutting federate is pretty fast.
 
Would not have to "switch" from DPM, but by adding inverse time functionality it would gain some compatibility with how most controls handle simultaneous rotary and linear motion.

Most all CAM systems will have a post outputting inverse time code that can be used to base a Brother post on. IME, not so many doing DPM. Leaves the Brother user having to figure out how to get their CAM post set up.
 
I very much doubt Brother will switch to inverse time from degrees per minute. The majority of the world works in metric and 100rpm in a cutting federate is pretty fast.

Very excited to get my new Speedio in a few weeks... but gotta be honest, I am really bummed that it doesn't support inverse time. DPM is old school/old tech. I am not sure I'd agree that most of the world works in metric... maybe you're right, but most of the USA is IPM and the USA is a HUGE market. I think it's most definitely a big hole in Brother's product line offering, and there's just no downside to allowing inverse time programming. It's just getting up to speed with how modern controls work.

Not trying to pile on - just adding my voice to the chorus on this one so you can pass that feedback on up the chain back to Brother and hopefully we might get a firmware update!
 
I think, as gkoenig pointed out, that it might be some BS Itar thing. Just like how their 5 axis machines won't do simultaneous 5x, only 3+2. It's not that Brother can't figure out how to do it, they're not allowed to export it.
 
Very excited to get my new Speedio in a few weeks... but gotta be honest, I am really bummed that it doesn't support inverse time. DPM is old school/old tech. I am not sure I'd agree that most of the world works in metric... maybe you're right, but most of the USA is IPM and the USA is a HUGE market. I think it's most definitely a big hole in Brother's product line offering, and there's just no downside to allowing inverse time programming. It's just getting up to speed with how modern controls work.

Not trying to pile on - just adding my voice to the chorus on this one so you can pass that feedback on up the chain back to Brother and hopefully we might get a firmware update!

Two points:

1- I’m hardly an expert on this subject, but I have my ear to the ground and I’ve never seen an MBT so dramatically improve a product via firmware over just a couple of years. If you go back and see folks original gripes with the first Speedios delivered? Almost all have been fixed or being fixed:

- High Accuracy Mode funkiness? Fixed
- No Mode B helical? Fixed
- Mode B 4th Axis? Fixed
- 2Mb program limit? Fixed
- G/M Code Macros in MDI? Being Fixed.

There is also a huge amount of undocumented/underdocumented capability onboard- Thermal Comp, Rotary Centerpoint Control, mysterious 3D surfacing tech for mass production. These machines are amazing because they are made (uniquely) by a company where 90% of the revenue comes from mass produced consumer and industrial stuff. They are just now realizing the Machine Tool niche of the business has done gone and built kind of an amazing general purpose machine. It’ll do nothing but improve, and almost all of those improvements are gonna be available in software.

2- The latest printed manuals indicate that GPS locks for ITAR compliance are on all Speedios and can be retrofitted. I suspect Brother’s long view sees full 5 axis as the future, but is laying the groundwork to implement it in a way that insures the whole business is not put at risk of ITAR’s absurd and capricious regulation. The thing many don’t realize is that an ITAR sanction against Brother would shut down the *entire* business (printers, sewing machines, IT products). The board is not going to put the enterprise at risk for a niche part of the business, hence what appares to be absurd limitations from our frame of reference.
 
Two points:

1- I’m hardly an expert on this subject, but I have my ear to the ground and I’ve never seen an MBT so dramatically improve a product via firmware over just a couple of years. If you go back and see folks original gripes with the first Speedios delivered? Almost all have been fixed or being fixed:

- High Accuracy Mode funkiness? Fixed
- No Mode B helical? Fixed
- Mode B 4th Axis? Fixed
- 2Mb program limit? Fixed
- G/M Code Macros in MDI? Being Fixed.

There is also a huge amount of undocumented/underdocumented capability onboard- Thermal Comp, Rotary Centerpoint Control, mysterious 3D surfacing tech for mass production. These machines are amazing because they are made (uniquely) by a company where 90% of the revenue comes from mass produced consumer and industrial stuff. They are just now realizing the Machine Tool niche of the business has done gone and built kind of an amazing general purpose machine. It’ll do nothing but improve, and almost all of those improvements are gonna be available in software.

2- The latest printed manuals indicate that GPS locks for ITAR compliance are on all Speedios and can be retrofitted. I suspect Brother’s long view sees full 5 axis as the future, but is laying the groundwork to implement it in a way that insures the whole business is not put at risk of ITAR’s absurd and capricious regulation. The thing many don’t realize is that an ITAR sanction against Brother would shut down the *entire* business (printers, sewing machines, IT products). The board is not going to put the enterprise at risk for a niche part of the business, hence what appares to be absurd limitations from our frame of reference.


Also, METI (Japanese export control) requires licensing for the products differently than the old TC products and depending on country, it is quite different than to the US. We used to not have to do anything really with Brother or METI on older Brother machines and now with Speedio, the restrictions are much more intrusive (from our point of view) but in line with most exports to other countries.
 
Bumping this for something I hadn't run into before. I'm getting feedrate alarms when my post processor spits out a linear (X) and Rotational (A) coordinate in the same line, with G01 and G94 active, and the feedrate is specified in DPM (which exceeds the IPM feedrate maximum of 787 in this control). Do I command IPM and let the C00 figure it out? In my particular case I've got a part where I'm cutting a radius that is located on the centerline of the 4th axis, so the radii cut fine it is the linking moves that are alarming out.
 
Bumping this for something I hadn't run into before. I'm getting feedrate alarms when my post processor spits out a linear (X) and Rotational (A) coordinate in the same line, with G01 and G94 active, and the feedrate is specified in DPM (which exceeds the IPM feedrate maximum of 787 in this control). Do I command IPM and let the C00 figure it out? In my particular case I've got a part where I'm cutting a radius that is located on the centerline of the 4th axis, so the radii cut fine it is the linking moves that are alarming out.


Ran into this issue a lot. Started using Inverse Time and changed User Parameter 193 from 0 to 1 to clamp at max feedrate instead of alarming.

The issue is its converting whatever feedrate (DPM or Inverse Time) in the backend to Degrees Per Minute (Can see this in the Position pages) and maxes at 1968 DPM. You can also only put in a F value at maximum 9999.999 which is nothing for Inverse Time. F values for Inverse Time are often 20x that. If there was a way to just get the machine to run in seconds instead of minutes the issue would for the most part be solved as the F values would be 60x smaller, but I haven't found a way to do that. This is why everyone says to run in Metric so the F values are 25.4x smaller. Even tried multiplying feeds by macro variables, which works, but still catches at the 1968 DPM.

There may be more to it on the back end but from everything I've tried, this is where I left it.
 
Last edited:
Ran into this issue a lot. Started using Inverse Time and changed User Parameter 193 from 0 to 1 to clamp at max feedrate instead of alarming.

The issue is its converting whatever feedrate (DPM or Inverse Time) in the backend to Degrees Per Minute (Can see this in the Position pages) and maxes at 1968 DPM. You can also only put in a F value at maximum 9999.999 which is nothing for Inverse Time. F values for Inverse Time are often 20x that. If there was a way to just get the machine to run in seconds instead of minutes the issue would for the most part be solved as the F values would be 60x smaller, but I haven't found a way to do that. This is why everyone says to run in Metric so the F values are 25.4x smaller. Even tried multiplying feeds by macro variables, which works, but still catches at the 1968 DPM.

There may be more to it on the back end but from everything I've tried, this is where I left it.

Having subsequently done inverse time feed based production on a Speedio, I can say that on the C-00, it is really intended for simple features and not for trying to mimic something like a 5 axis simultanious swarf cut for a Class A surface finish. J hook feature on a cylinder? Yes. Relief cut on a slot? Yes. Swoopy swarf surface on a consumer product? No.

The D-00 is a totally different animal. The limits on the C-00 had to do with the encoder resolution and processing power. They have gone from 1.2 Million pulse encoders to ~ 10 Million on the D machines (all linear axes and any rotary). Inverse time now works perfectly as the limit went from 4 significant digits (9999.999) to 6 (999999.9999), and the M200 has full 5 axis simultaneous motion with Tool Centerpoint Control. On a Brother T200Ad table, with full parameter and acc/dec configuration from the factory, any D-00 machine can run intricate inverse time feed cuts just fine, and any machine with the D-00v (5Ax) control in the future is going to be a monster.
 
Well I also discovered that I had messed up something in my post processor as well. I ended up rebuilding that with the most current base post from Autodesk and now I can run a lot faster in the standard DPM/FPM hybrid mode. I ended up just indexing for my highfeed roughing toolpath and that way I can run it at normal speeds and feeds.

The M200xd1-5AX looked quite capable at IMTS. For sure the D00 and it's more standard handling of things like tilted work plane and tool center point control are game changers. When the Yamazen tech told me I needed to switch to metric for inverse time I was like "isn't the whole point that it doesn't care about units?" The encoder thing makes a lot more sense.

Now if only they had been really making that sucker dance at the show. It was going slow as hell cutting air LOL.
 
When the Yamazen tech told me I needed to switch to metric for inverse time I was like "isn't the whole point that it doesn't care about units?" The encoder thing makes a lot more sense.

You can do inverse time feed in the C-00 in inch, but if you do the math by hand on an inverse time feed cut, you find that the F numbers in metric are an order of magnitude smaller than the same F numbers for the same cut in inch. Given that the C-00 is limited to 4 significant digits for F numbers, switching to metric dramatically improves this limitation. Not so relevant now with the D-00 as 6 significant digits is more than enough for any application in inch or metric.

This is why I think the C-00 is fine for simple 4 sim axis features like a J hook or scallop cut which might only be 2-3 lines of code with long enough travel that the F numbers are small (longer travel length = shorter F number in Christopher Nolan's inverse time world). Trying to trace a nice spline profile requires very short moves, which drive the F numbers way way up.

Now if only they had been really making that sucker dance at the show. It was going slow as hell cutting air LOL.

All of the IMTS setups were done by Brother and the same ones on the Brother Machine Tools YouTube channel. They also sent 2 of the 5 axis engineers who cranked up the feed rates a few times and demonstrated that she does, in fact, chooch pretty good. Lots more 5 axis demos are coming... Nobody was really excited by the M200 5Ax at first because we all thought the price would be astronomical and it would be a niche machine, but the pricing turns out to be pretty incredible for a full 5 axis mill with turning capabilities and a low-cost integrated automation option.

Honestly, if I was going to start a job shop as a more seasoned fellow now, I would buy 2-3 M200s with BV7s and go tackle the medical market making implants. Which is 100% what Brother intended this machine for, and I think they nailed it.
 








 
Back
Top