Shapeoko/computer disconnect as soon as spindle is turned on

Hello. On this episode of problems I’ve encounter while trying to use a spindle:

So I got my replacement spindle and vfd in. Last one cooked the vfd. This is an HY inverter and a matching 2.2kw spindle from amazon.

I have everything hooked up. Up to the vfd was planned permanent install. My vfd to spindle wires and coolant wires have not been cleaned up yet. They are hanging over head in the enclosure. I have larger tracks that need to be installed to hold everything that way but wanted to check functionality first, so far I can have the cnc operate or the spindle operate not both lol.

The vfd is being controlled manually for the time being. I am starting and stoping the spindle and setting rpm via the vfd controls.

I can start the machine, initialize it, it will go through bit setter motions and return from center like it should. The second I start the spindle the computer disconnects from the cnc. In order to reconnect I have to cut power to the cnc board and turn it back on.

Things that have changed since the last short lived spindle a couple weeks ago:
Everything moved into an enclosure
Hdz upgrade,
Motion stop switch installed.
I broke the usb cord that comes with it. The replacement does not have suppressors on it- it shouldn’t be so difficult to find one that does have them but it is.

I had this setup operating with the carbide router just fine.

I installed the new vfd and spindle with carbide mount today.

Any ideas? I’m assuming it’s an interference issue but I’m not sure.

Just wanted to update that I’ve been ruling what I can out. It does not do this when using my laptop

Usually the Chinese spindles are not grounded. Did you run a ground wire to the spindle body?.

Are you saying you get no disconnects when driving the Shapeoko from a laptop and starting the VFD, but you do get disconnects when using another computer ?
If so, you could try and get a USB isolator (and while waiting to receive it, start with trying a different USB port on the same computer)

Does it also happen if you start the spindle at minimal speed ?

That is good to hear after all the trouble you had with the last one

That’s a nice enclosure and some pretty serious wiring, I like it.

Do you have a power filter installed on the input feed to the HuanYang inverter? The reseller in the UK I got mine from was very clear with me that HY inverters are particularly noisy and that his customers either bought an input power filter with the drive, or a few days after installing the drive. I paid the extra £15 and wired the filter in as part of the install.

I concur that there’s probably interference, either from some differential in the ground or from radiated EMI pickup.

The ferrites on the USB are quite important to keeping out RFI.

Is your spindle cable screened and grounded?

Some folks have had mileage out of putting some ferrite rings on the outfeed cores of the VFD to quell some of the RFI at source.

Lot of question to answer.

As far as the wiring on the enclosure. The idea is the entire cnc setup ( shapeoko, spindle, lighting, dedicated shop vac, chiller,computer everything) is all powered off a single 220v plug and distributed from its own breaker box. This way I can unplug the plug(50amp) and move the enclosure around( it’s on wheels should I choose to do so. If for nothing else it makes cleaning easier. Also will make moving easier in a couple years.

As far as the replacement spindle I went through a different seller to get a replacement. AliExpress is making whip the old one back to China for a full refund including original shipping costs but the return shipping is on my. I actually like the g penny spindle build quality a lot better but I wasn’t going to get bit twice. Went from a 1.5 kw to a 2.2kw while I was at it as it was shipped from in the USA.

Vfd to spindle wires are not shielded. I did wrap on foil tape and ground the foil at the spindle to see if it helps as a last ditch effort and it made no change.

In like the g penny spindle this one was not grounded. I opened it up and added a grounding wire to resolve this.

I have a shorter usb with some ferrite cores at my brothers I need to go pick up.

Last night I was able to grab a laptop and initialize and turn the spindle on without issue. That’s as far as I went. I’m going to try and see if any of the other USB ports work on the tower I’m using but I’m pretty sure they don’t. I think that’s one of the reason we replaced it

I’m not sure you saw my response to the facebook post, but I would advise you to check the initial start frequency. Most folk leave this as 0, however the spindle/vdf can trigger a short burst with disrupts things on the same circuit. It has been known to trigger a MCB.

If you set the minimum run frequency to around 133hz or 8000 rpm it might fix the issue - do this on the VFD.

1 Like

Was that the full length of the spindle power cable?

Hmmm, electrical safety is obviously not the highest priority for these manufacturers.

If you don’t have an input filter on the VFD, here’s the one I got, they’re pretty generic, you’ll see this has a diagram of it’s simple LC filter on the sticker. Withoug a filter the VFD is going to inject lots of noise back into your power cabling and use all of that as a broadcast antenna. That could well upset the tower PC and not a laptop.

Yes the foil was the entire length

I will see if I can source an input filter locally.

Also I’m not sure where the feed back is coming from. I’m the corn too box I have a relay that’s powered on by my magnetic start/stop switch. This relay controls both hots on the 220v to vfd and chiller and it also controls the hot to the 110v circuit for the shapeoko power supply. So that’s another spot it could be getting emi

I did see the reply to my Facebook post this morning. Will try that when I get home. I’m at work for 7 more hours so it will be a bit.

Wanted to update on here. Long day at work with another one tomorrow but for a bit to try things.

Long story short I moved the computer tower to the opposite side of the enclosure from The vfd and all the electrical. And it seems to be fixed although I can’t run any files tonight. I will need to reroute some wires so I don’t have interference with my new larger wire tracks.

I’ll mess with it on Sunday. Tomorrow is a busy day. The joys of 4x 12 hour shifts

2 Likes

Well had some time to work on this some more. The movement of the computer doesn’t fix it completely but it helps.

If i let the router start at a low speed it looses connection.

If I start it at 10k rpm ( my spin up time is 2 seconds to 24k) most of the time I can get through without a disconnect.

Then if I start moving the machine around using rapid position i will get a disconnect after a couple moves.

Ferrite cores will be here today. The power filter will be here tomorrow.

2 Likes

Well the ferrite cores made it better. Still can’t get it to the point it starts cutting. Seems to be disconnecting on the original z plunge before it hits material.

So the power filter will be here today now according to amazon. It just goes on the the two hot legs of the vfd power correct?

That’s probably progress, there’s probably interference from multiple sources so there’s a few moles to whack.

Do you have a link for the power filter so I can take a look pls? I might say really useless things otherwise :wink:

https://www.amazon.com/gp/aw/d/B07FDYRQRY?psc=1&ref=ppx_pop_mob_b_asin_title

Just showed up

1 Like

2 Likes

That looks like a pretty good filter, it’s multi-stage (two sets of series inductors followed by capacitors across the lines) so it should do a pretty good job, mine is only single stage and seems fine, nice find.

Wiring it up, if you’re going to be putting this in your big metal wiring cabinet, then it should be pretty easy, basically, yes, in line with the power feed to the VFD but it needs the ground terminal to the local earth in that cabinet as it filters to ground as well as between the lines.

There’s mine on the right on the little vertical bracket in line with the output cable taking power to the VFD.

1 Like

Had the same issue initially and found I miswired ACM and DCM

My VFD manual states the spindle should have a ground separate from everything else. Easy thing to try.

Also there are spindle carrier frequency adjustments that can effect motor noise and electrical interference. Not sure if the HY has that option though.

3 Likes

Good point, yes it does, I turned mine all the way up to 15 initially because my spindle sang like a drunk Welshman, it quietened down after some running in.

If your HuanYang is the same as mine, it’s this setting;

None of that is currently hooked up. Spindle rpm is being controlled manually until I figure this out.