HY VFD and spindle at low speed

I've been a bit adventurous recently and experimented with cutting steel on my router (do a search for AVOR in this forum to find a description). The machine was originally intended for wood-cutting with maybe a bit of aluminium but I was in a "what the hell" mood and tried some bright mild steel. Basically, a project calls for a kind of 3D jigsaw in 3mm steel plate, silver-soldered together. The original designer called for tapped holes into the edges of the plates to take temporary bolts, which sounded a bit fraught and a lot of work. I could have bought laser-cut plates but they cost, and I wanted to keep the project in-house, so to speak. This is a hobby activity, not commercial, which moves the goalposts a bit - I don't cost my time, within reason. So I drew up the structure in Fusion 360 with a tab-and-slot design to make it all self-jigging, and set it all up on the router to see what happened.

Problem is cutter speed. After trying a standard endmill (2.5mm), which worked but looked fragile, I bought a couple of 2.5mm 3-flute carbide coated cutters. Only 5mm flute length and on a 6mm shank, so much stiffer. Recommended for dry cutting of steel (which suits me - keeps the machine cleaner). However, according to the data sheet these cutters should be run at around 4600RPM. I've been cutting fairly successfully like this, playing with depth of cut starting with "very conservative", but this is really below the speed at which my spindle should be run - I know from experience that torque is dropping off fairly quickly at these low speeds. Curiously, I seem to be suffering from small surges of speed/torque(?) from the spindle - at fairly regular intervals of a few seconds there is a slight "thump" from the spindle, a tiny bit of extra vibration from the cutter, and then it calms down until the next time. I'm wondering if I'm running below the speed at which the torque feedback or speed control or whatever mechanism is used is effective. I'm slightly increasing DoC each time, to put a touch more load on the spindle (although I suspect that I am really not needing much torque to turn these tiny cutters) and I have played a little with the cutter speed - increasing spindle speed seems to help although I don't want to go too far over the cutter manufacturer's recommendation.

Does anyone else have any experience of the typical HY VFD/spindle combination at lower than usual spindle speeds? Anyone played with any of the tuning parameters to aid performance at low speeds? I'm also looking through the HY manual to see if there's anything that might help my understanding, but decoding Chinglish at the same time doesn't make it easy!

BTW, it is a water-cooled spindle, so no overheating problems at low speed.

Quote:

---

Originally Posted by Neale

I've been a bit adventurous recently and experimented with cutting steel on my router (do a search for AVOR in this forum to find a description). The machine was originally intended for wood-cutting with maybe a bit of aluminium but I was in a "what the hell" mood and tried some bright mild steel. Basically, a project calls for a kind of 3D jigsaw in 3mm steel plate, silver-soldered together. The original designer called for tapped holes into the edges of the plates to take temporary bolts, which sounded a bit fraught and a lot of work. I could have bought laser-cut plates but they cost, and I wanted to keep the project in-house, so to speak. This is a hobby activity, not commercial, which moves the goalposts a bit - I don't cost my time, within reason. So I drew up the structure in Fusion 360 with a tab-and-slot design to make it all self-jigging, and set it all up on the router to see what happened.

Problem is cutter speed. After trying a standard endmill (2.5mm), which worked but looked fragile, I bought a couple of 2.5mm 3-flute carbide coated cutters. Only 5mm flute length and on a 6mm shank, so much stiffer. Recommended for dry cutting of steel (which suits me - keeps the machine cleaner). However, according to the data sheet these cutters should be run at around 4600RPM. I've been cutting fairly successfully like this, playing with depth of cut starting with "very conservative", but this is really below the speed at which my spindle should be run - I know from experience that torque is dropping off fairly quickly at these low speeds. Curiously, I seem to be suffering from small surges of speed/torque(?) from the spindle - at fairly regular intervals of a few seconds there is a slight "thump" from the spindle, a tiny bit of extra vibration from the cutter, and then it calms down until the next time. I'm wondering if I'm running below the speed at which the torque feedback or speed control or whatever mechanism is used is effective. I'm slightly increasing DoC each time, to put a touch more load on the spindle (although I suspect that I am really not needing much torque to turn these tiny cutters) and I have played a little with the cutter speed - increasing spindle speed seems to help although I don't want to go too far over the cutter manufacturer's recommendation.

Does anyone else have any experience of the typical HY VFD/spindle combination at lower than usual spindle speeds? Anyone played with any of the tuning parameters to aid performance at low speeds? I'm also looking through the HY manual to see if there's anything that might help my understanding, but decoding Chinglish at the same time doesn't make it easy!

BTW, it is a water-cooled spindle, so no overheating problems at low speed.

---

Hi Neale,


Lower than normal speeds.

Yes i had a bad experience actually.

When i was trying to get 0-10volt working i had my spindle running at a low spees to long. maybe 10 sec.


Things heated up quickly.

I hope i have no damage.

I made sure minimum speed is 50hz. at 0 volts.

Hope this helps

Gerz Bert.

Verstuurd vanaf mijn SM-A320FL met Tapatalk

Bert - thanks for the comments.

I'm not worried about spindle overheating as I'm only a small amount below its "official" minimum speed and the water-cooling keeps the spindle body at more-or-less ambient temperature even after 30-40 mins running. However, there are quite a few VFD settings which do things like boost voltage at lower speeds, and there are some torque-sensing features that might do something similar. However, if 400Hz is 24K RPM, then 5K RPM is still over 70Hz which seems to be reasonable for the VFD. I do need to have a look at the "torque" control settings on the VFD.

The effect I'm seeing is a bit like a governor on an old gas engine - the speed slowly drops until the governor detects that it has fallen too far, then there's a bit of a surge to bring the speed up again. In my case, I can't really detect speed changes, just this kind of "thump" which is a bit like a scaled-down motor-start effect. The VFD readout flickers between two similar readings, but it always does that anyway at any speed, and the difference between readings is very small. The speed displayed on the VFD is within a couple of per cent of the Mach3-demanded speed, which again is normal on my machine.

I'm using the router and spindle for something that neither was really intended to do so I wasn't really expecting anyone else to have experienced this. But it seemed to be worth asking the question!

Torque doesn't decrease with speed, Power does. Torque remains constant (at least in an ideal world, but there may be a slight variance due to things never being ideal).
And torque is directly proportional to current.

The only way you're likely to kill an induction motor with adequate cooling at rated speeds, is by pushing too much current through the windings, which can cause them to overheat. Even though the spindle is water cooled, heat transfer won't be ideal, so by pushing too much current, you can still overheat the windings.
Torque boost options simply increase the current. For short boosts it's not likely to be a problem, but prolonged it is.

For this kind of application, I would be monitoring the motor current, as that will give you the best indication of just how much load you're putting on the motor.

I agree as far as the motor is concerned, but the VFD has various torque-related settings and corresponding parameters, including things like voltage-frequency linking which can boost torque at lower speeds by increasing applied voltage - useful during acceleration if the motor has load on it at startup. I think - I'm still trying to understand the manual and the various settings available. I'm also aware that there must be some kind of feedback system as there is a set of PID parameters that can be modified, which implies feedback. Given that the speed is notionally determined by VFD output frequency, modified to some small extent by slip, I assume that this feedback system will boost applied voltage and hence current to increase torque if the sensing system believes that the motor speed is too low due to applied load on the spindle. Maybe - it's guesswork on my part at this stage!

One thing that does seem to be true is that increasing depth of cut and hence spindle load by a small amount (0.4mm to 0.5mm) seems to smooth things out slightly, which agrees with my thinking that any feedback mechanism in place, governed by the PID parameters, is tuned on the basis of a more heavily-loaded motor than I am presenting, or at least the parameters are not really optimised for the speed/load combination I am using.

I know that the display can be switched to display motor current, so I'm going to have a look at that tomorrow, running the motor at various speeds. Assuming that the displayed value is actually an accurate representation of motor current!

Trouble is, the rate of cut is low enough that I am standing there vacuuming up chips for long enough that my mind starts speculating on what's going on; I need to keep my "engineer" hat on long enough to just get on with it while it's working!

Fossilised somewhere on this forum (4-5 yrs ago) there's a thread/post where I've discussed how to get more torque from the common 2.2kW spindle and HY VFD combination at low speed. Essentially you use the voltage vs frequency mapping settings to boost the voltage applied to the spindle at low speed. This will put more current through the stator windings, so in my case I had something like 2-3 amps read by the VFD at just 4800rpm, but it did give me enough extra torque to drill aluminium with a 4.2mm HSS drill at that speed.

Adding extra heat in the stator is not too big a deal - in theory you could go all the way to the current rating of the winding, and a bit more as the iron losses should stay fairly small at low speed. However this will induce extra current in the rotor, which is much harder to get the heat out as it's not directly water cooled. I did once characterise the 2.2kW spindle to find the equivalent circuit, so you can work out how much current is permissible in the stator to have the same rotor losses as it was designed for. How much torque you actually get out of this is harder to say though, as we don't know the saturation characteristic of the stator. Not sure where that spreadsheet is now though.... so I suggest you increase the voltage until you see a few amps with no load and see if it's enough.

Another approach is to use a cutter with the right coating to enable raising the spindle speed. These have worked well for me:
https://www.shop-apt.co.uk/carbide-e...ted-65hrc.html

Thanks, Jonathan. I'll have a search back through forum history - I hadn't remembered this one being discussed before. My own background is electronics rather than electrics so while I understand the basics, I'm a bit hazy as to exactly what happens inside these motors. However, my thinking had also been to tweak the VF curve at the bottom end - thanks for the suggestion of using motor current to get a bit of insight.

This is the first time I have ever seriously used carbide milling cutters. I've generally used HSS on the mill and carbide for routing wood but I've been driven to carbide cutters because of this minimum spindle speed issue. I bought the ones I am using after a quick bit of web browsing but along with the cutters I was sent an enormous catalogue and looking at that, I see that there are indeed cutters that might be more suitable - at a slightly higher price... My concern about those is breaking them, either due to running them into a clamp (it happens!) or vibration/resonance due to a less than optimally-rigid structure. Ok, my machine is all welded steel with profile rails and ballscrews, but I'm still surprised-but-pleased that I can cut steel at all.

Those APT cutters look like a good deal so I'll do a bit more research down that line as well.

I am very interested if you guys find out we can use settings to boost low rpm torque without burning up the motor.

Do we have a reason to think HY put unreasonable reservations in the settings ... over protective measures so to speak?

I would like low rpm performance too.

Grtz Bert.




Verstuurd vanaf mijn SM-A320FL met Tapatalk

Neale - Thank you, very interesting thread. I hope you will post your results please. I have been down this road on a similar router with moderate success using carbide+coated cutters. I did find a definite advantage adding just a smear of the blue EP metal working lubricant, with a finger, to the projected toolpath. So minimal mess and it also kept some of the chips on the metal surface.
Your VFD setting results would be particularly welcome as have to admit a bit windy of touching my working settings.

I'm afraid, guys, that I'm probably not going to be experimenting with VFD settings, at least not for my current project.

I wasn't sure that there was a lack of torque - pushing a 2.5mm cutter, even through steel, isn't going to need very much - but rather a problem with speed control at low speeds. Watching the cut in progress, I suddenly remembered what the effect reminded me of. I was seeing what was almost a tiny pause, a slight release of the cutter, and associated squealing and poor finish. Just like turning when you take a shallow cut on a metal prone to work-hardening. So I wound up the feed speed a little which helped significantly. I then ran out of the cold-rolled bright mild steel I had been using and swapped to hot-rolled black steel. Which cut much better and without the vibration/"pausing" effect - so I think I probably found the answer. I suspect that the bright mild knocked the edge of my cutter, though, as it didn't last much longer even in the softer metal. I'll ramp up DOC next time and see how it goes.