[ihavenofish]

For steel you need to choose the right cutter and strategy. tialn coating is going to be mandatory, as will an adaptive clearing style of cutting. You are definitely going to be restricted to very small cutters.

Using the sandvik calculator with the specs of the jianken 40000rpm 750w spindle, It gives a number of plausible recommendations for cutting 4140 with 1/8" bits at 15-20krpm. Not ideal, not fast, not rigid, but for the odd steel item (were talking routers here generally, not mills) it will work. If you cut steel often, or have a mill, then yeah, this is not the type of spindle you want at all.

[machinedude]

Hi,
as I posted earlier I have a 750W 24000 rpm spindle. In my ignorance I thought it would cut steel if I took light enough cuts.
My optimism was misplaced. I did manage to cut steel by slowing the spindle to 9000 rpm and using a 3mm diameter tool
and very VERY light cuts. The slightest mistake with your programming or feed rate and the spindle would stall and break the tool.

I was rather disappointed, but now realize that my expectation was TOTALLY unrealistic.

What I did was make my own spindle based on an Allen Bradley 1.8kW 6Nm (cont) AC servo. I bought a Rego-Fix ER25 cylindrical
tool holder, fitted matched P4 NSK angular contact bearings in my own housing. I made it so it fits the same clamp as my highspeed spindle.
Note that I direct coupled the servo and tool holder so 3500 rpm is max.

It works really well in steel and stainless with tools up to 16mm diameter up to the limit of the rigidity of my mill.

I would have to recommend either a direct or belt coupled servo motor for a spindle IF you require high torque for cutting steel.
My experience is that the highseed spindles, at least those that hobbyists can power at home, say less than 2.2kW, are not up to
cutting steel.

Craig

the thing you have to remember when cutting steel is speeds and feeds and for that type of material. 9,000 rpm's is no where near slow enough for cutting steel. A general rule of thumb is the smaller the cutter the faster the speed you can run but your still not going to get into the range of speed needed with these high speed spindles. the place where the high speed spindle will do well is with aluminum for the simple fact with the high speed you can get a very high feed rate taking very small cuts and remove the material at a decent rate.With CAM these days you can get away with a lot with some fancy tool paths. i would not get to hung up on depth cuts to much and use all the flute with a small step over to control cutting forces. it's good practice to not over use the tips of the cutters and use as much of the flute to do the work for material removal.it would not hurt to have a coolant mister as well not only does it help finishes but running higher speeds in aluminum with set the tool up for galling and a mister will prevent this and also clear the chips for the cutting path.

a mister would help with steel too since chances are your running to fast to begin with, so the chips are going to be hot as a result and the mister would get you a little better tool life, but it's not going to fix anything long term just slow the process of the cutter burning up down some. and i am only talking about mild steel here, i would save you tool and machine if it were something like stainless steel.

[joeavaerage]

Hi,

the thing you have to remember when cutting steel is speeds and feeds and for that type of material. 9,000 rpm's is no where near slow enough for cutting steel.

I quite agree. It was just that 9000 rpm was a slow as I was prepared to go without my air-cooled spindle overheating. I did do a few jobs with 3mm tools but
it was slow. Additionally as you point out I was only ever using the tips of the tool, and so while the tool dulled the vast majority of was untouched and I was throwing
them away. It was then that I decided to make a spindle that could do a good job in steel and stainless with both HSS and carbide tools......that means
low rpm and high torque.

That is what I did. It takes about 5 minutes to swap from the asynchronous spindle to the servo driven one, so that's what I do. The other thing that I have added
is flood cooling....its great. One of the most important aspects of cutting metals, any metals, is to remove the chips from the cutzone.......re-cutting chips
is the death knell for ANY milling job.

Craig

[machinedude]

with carbide your in the ballpark with speed but carbide does not like to vibrate due to lack of rigidity so that can cause problems for you, since your stalling the low end speed probably lacks the torque witch I've noticed others mention so it seems pretty common. and when you get into the small cutters they can't remove a lot of material fast just because they just don't have the strength of a larger cutter. it actually makes a big difference on how much of the cutter is sticking out of the holder when they get that small or smaller. when using a small tool it's best to pick one just long enough to do what you need and set it in the holder so you have just enough clearance.

[machinedude]

yeah low speeds and flood coolant are best so your on track there now it seems. every case is different based on the machine itself. they all have there own personality it seems and what works on one does not always work on another. a lot of it depends on how rigid the machine is and spindle speed ranges. i plan on poking my head down the rabbit hole and playing around with mini mill head on the gantry mill / router i am working on. I don't think flood coolant is an option for me but have a mister to help out. nothing ventured is nothing gained the way i look at it.

[joeavaerage]

Hi,
that's the whole point of building my second spindle. My first spindle (750W 24000rpm) is great for engraving, making PCBs, and I do a lot of those,
and aluminum up to about 6-8mm tools. If I want to cut steel then I fit my second spindle (1.8kW (cont) 3500 rpm 6.2Nm (cont))......no shagging around
with tiny tools and worrying about stickout or low cross section tools or whether my Gcode is too aggressive......just load a 16mm carbide tool, turn on the
coolant and watch the chips fly.

I cut 304 and 316 no problems. The hardest/toughest steel I cut is hardened and tempered (1.0 GPa) 4340. You have to concentrate on the tough stuff,
if the cooling is not right, or the biggest sin, re-cutting chips, then you will have a disaster.

My current mill, really a mini-mill (machining volume 180mm x 180mm x 180mm) has Vexta 5phase steppers and low lash (<2 arc min) 10:1 planetary
gearboxes. The reduction means I get great torque and consequently acceleration and thrust but at the expense of speed, G0 rapids are 1200mm/min.
I thought it not really fast enough to consider the tricky trichoidal and similar toolpaths required by the 'High Speed Machining' strategies propounded
by the CAM manufacturers. Aside from the fact that I just about need a mortgage to buy a CBN tool!

I am currently building a new mill, still small, 350mm x 350mm x350mm machining volume, but it will have direct coupled AC servos which will
allow G0 rapids of 25m/min and max G1's of 15m/min. I will have to give some serious thought to exploring HSM toolpaths. In the mean time it
will be many times more rigid than my current mill so I'll be able to harness all of the power of my second spindle.

Craig

[machinedude]

small mills are a heck of a lot easier to make rigid they just lack weight. what i have going is probably just crazy for a 4' x 8' gantry table but time will tell i guess. it will cut it's just to be determined how well it cuts. i have another build i started but this kinda took over since if it works it will help with my other build and speed things up. my other mill build is more VMC style and i have no doubts about it being solid. for that one i have a spindle started and have a 750W DMM servo to drive it along with 3 more of them on each axis. servo controllers are expensive to build.

i have some pretty large bearings in that spindle and built it are around a BT30 taper. the housing and shaft are made out of 4140 HT not small or light in weight for sure. close to 50 lbs actually.and it's not cheap to build one yourself since this is on topic for this tread.

[ubergeekseven]

Ihavenofish points out an important point with respect to much of the Chinese produced machinery. That is specs for horsepower are often input power to the system. In some cases it amounts to false advertising as the motors can’t even sustain those numbers long befor burning up. However even if the power numbers honestly reflect power input under load conditions it isn’t what is expected in the west (USA anyways) where we expect to be able to get that power at the output shaft. I’ve actually have been burnt by this buying other shop tools.

Now that being said I do believe most of the Chinese spindles are accurately rated, at least better than some of the other stuff coming from China. The problem is often the user not understanding the torque - speed relationship to get those power numbers. In simple terms you don’t need a lot of torque at 25000 or 30000 RPM to output 2 KW of energy. Torque however is very important as you try to lower the spindle speed to something suitable for the work in progress. As someone noted above large diameter cutters can be a problem as you lower the spindle speed. To really understand how a spindle will perform you need the torque and power curves. Beyond that cooling is a huge issue as the RPMs slow down. This especially if one tries to push the spindle beyond its capabilities.

I’m actually surprised that there isn’t a greater use of belt driven spindles on home builds. It is a fairly cheap way to have the option of torque at low RPMs and yet be able to obtain reasonable routing RPMs. Belt drives are perhaps the cheapest way to a wide range of useable spindle RPMs.

I had no idea, until this thread, the lack of power behind the 2.2kw spindles. I have the cncrp 4896 pro and didn't get the spindle. I have, several times, thought about purchasing it. They specifically told me that the Bosch 1617 can only produce the specified power by running at full speed and that their spindle has the rated horse power from top to bottom speeds.
Is there something about their spindle that is different from the other 2.2kw air cooled spindles from these other manufacturers? Is their spindle actually capable of this claim? It's a huge factor that I had no idea to weigh my decision. I mainly do wood. I have a 4th axis. I've machined brass, aluminum and cast acrylic with good results after playing enough. Will a 2.2kw spindle allow me to do what I do now better? I was hoping for better control with aluminum parts.

Sent from my SM-N960U using Tapatalk

[ihavenofish]

I think their wording is misleading. Their spindle (should) have constant torque from 8000-24000rpm. It is unlikely to have constant power in that range - although there are some spindle built like that. Usually large milling ones.

Power is torque and rpm together. So, if you had constant torque, like almost all of these basic induction spindles, you have 2.2kw at 24000rpm, 1.1kw at 12000rpm, 550w at 6000rpm and so on.

teknomotors makes some spindles that have constant POWER from 12000rpm to 24000rpm. They have 750w at 12000rpm, and 750w at 24000rpm etc. This means they have twice the torque at 12000rpm as they do at 24000rpm. My teknomotor atc spindle is the same - 12000rpm 3.7kw, 24000rpm 3.7kw, 3000rpm 925w.

[ihavenofish]

Heres 2 tekno spindles to compare

2.2kw, from 18000-24000
https://www.damencnc.com/en/electros...00rpm/a14?c=32

1.8kw from 12000-24000
https://www.damencnc.com/en/electros...00rpm/a17?c=32

The 1.8kw is the longer motor, with more torque at low speeds, less torque at higher speed.

Those ratings are output, continuous of course. So if you do metal, especially steel, the 1.8 tekno is better than the 2.2, and far better than the cncrp 2.2 assuming that one is rated to 24000 and as input (like all the other chinse spindles, they don't seem to ever provide any data).

Of course, that tekno is $1000, so, you have to pay to play :P

An exception might be this one:
https://www.damencnc.com/en/electros...00rpm/a32?c=31

Which is the same 2.2kw model, with an electric fan. This will make running in the low speed ranges more reliable and safe, so, if you always ran at 6000rpm for some reason, you might forgo the extra torque of the 1.8 to get the reliability of fan cooling.

[ihavenofish]

Ha, I just noticed the price of the cncrp spindle.... Sometimes I don't understand companies...

[joeavaerage]

Hi,
Mechatron do a similar sort of thing, this is a 4kW model, it has constant torque (4Nm) from 3000 to 9000 rpm and constant power (4kW) thereafter to 18000 rpm.
Note that it has three pole pairs and requires a VFD capable of 900 Hz output to achieve all these things. I believe this spindle is several thousand Euros so
could not be called budget either.

https://www.mechatron-gmbh.de/filead...FP-M-12040.pdf

[ubergeekseven]

Heres 2 tekno spindles to compare

2.2kw, from 18000-24000
https://www.damencnc.com/en/electros...00rpm/a14?c=32

1.8kw from 12000-24000
https://www.damencnc.com/en/electros...00rpm/a17?c=32

The 1.8kw is the longer motor, with more torque at low speeds, less torque at higher speed.

Those ratings are output, continuous of course. So if you do metal, especially steel, the 1.8 tekno is better than the 2.2, and far better than the cncrp 2.2 assuming that one is rated to 24000 and as input (like all the other chinse spindles, they don't seem to ever provide any data).

Of course, that tekno is $1000, so, you have to pay to play

An exception might be this one:
https://www.damencnc.com/en/electros...00rpm/a32?c=31

Which is the same 2.2kw model, with an electric fan. This will make running in the low speed ranges more reliable and safe, so, if you always ran at 6000rpm for some reason, you might forgo the extra torque of the 1.8 to get the reliability of fan cooling.

The cncrp package is about $1800 when getting their vfd and kit to connect it prewired. So that tekno, with vfd may be less.

Sent from my SM-N960U using Tapatalk

[ihavenofish]

The cncrp package is about $1800 when getting their vfd and kit to connect it prewired. So that tekno, with vfd may be less.

Sent from my SM-N960U using Tapatalk

It is definitely less, and damen offer prewiring as well.

I'm kinda irritated by that haha.

[kmagyoyo]

Let’s remember this is the DIY router section of the forum. Routers are typically used at higher feed rates with non-ferrous metals, wood, composites, or plastics. If you’re milling stainless and 4130 then I think we can all agree that a router is poorly suited to the task in terms of rigidity.

My former router had an HSD 4.7 Kw spindle that had no trouble plowing through 3/4” plywood with a 3/8” 2-flute compression bit at 450 IPM. That sort of performance was necessary in the commercial environment but now that I’m strictly hobbyist, what’s a good real-world power rating to be able to do similar work but without the time constraints (e.g., partial-depth passes)?

[joeavaerage]

Hi,

Routers are typically used at higher feed rates with non-ferrous metals, wood, composites, or plastics

Agreed, and given that ferrous metals are OFF the table then any of the cheap Chinese 2.2kW spindles will work for you.
There are so many advertised that its impossible to indentify one as better or worse than another. There was a thread recently that
I read that the poster had noted one manufacturer, Jiang, or some name like it, made a respectable spindle and was in fact the original
manufacturer of many of the spindles advertised but under a different name.

I will try to track the post down and link it to you. It is the only post that I recall that had a specific manufacturer recommended.
I would avoid the cheapest Chinese VFDs. I have used Delta (Taiwanese manufactured in China) and am very happy with them.
They are not the cheapest VFD, but along way cheaper than many Japanese, European or US made VFDs.

Craig

[ihavenofish]

I have used 750w tekno/colombo spindles on my small routers (xzero), and have rarely been wanting for more power. You need a pretty beefy machine to make good use of more power anyway. I cut hardwoods, not plywood, but that spindle would easily cut with a 3/8 bit 3/4" depth in mahogany at 50-100 ipm. Generally I took lighter passes due to fixturing limits and trying not to blow out chips in the wood.

[v2tiller]

I also have a 750w teknomotor spindle with delta vfd. I got the model with max power at 12000 to 24000 rpm. It also has a bearing arrangement for drilling as well as side loads and er20 collet size.
So far i love it and im only cutting ally. I've used 8-10mm drill bits and 10mm end mills in it and it works great.
My diy router is mostly a steel build though so quite solid.
Im sure at some stage ill want more power, then atc and then a real vmc!
But until then im happy i paid a bit extra to get a good quality spindle.

[ihavenofish]

Another interesting option
https://www.hsdusa.com/viewdoc.asp?co_id=1075

HSD 30krpm 1kw water cooled block spindle with an air purge. Price should be in around $1000us.