VFD error

[routercnc]

I was cutting this morning and all of a sudden there was a large blue flash from the connector into the top of the spindle, and the RCD tripped on the consumer unit.

The connector is the standard 3 pin type for chinese wc spindles. Looking inside the female part of the connector (from the VFD) showed a burn mark around the clear plastic sleeve. I checked the resistance between each socket and earth, checked the spindle pins for resistance (about 2.2ohms so ok there) and resistance to earth. All fine.

I then noticed that the pins which stick up inside the spindle connector are split down the middle to provide some spring loading and make good connection. 2 of them had a gap between, but the third one was fairly close together. This meant that when the connector was applied, the pin wasn't making a great contact.

I think this was enough to momentarily break the circuit to the spindle (while it was cutting!). Part of the reason for this is probably because (like many DIY machines I suspect) the cable from the VFD goes into the spindle and any movement of the machine during cutting continually strains the cable. I've been running like this for about 4 years, and kept meaning to make a strain relief arm, but never have done.

Knocked one up in about 30 minutes after the problem, but it's too late now.

Now I can't start the VFD. It keeps telling me there is a fault (E14 - UVW to earth fault). It does this without the spindle connected, plus without the VFD cable connected. There is a reset procedure which involves briefly connecting a terminal mark P24 (24v?) to pin 5, which make symbols on the display goes round in little circles. But the error remains.

I've tried switching on and off many times, held the reset for a long time, nothing. Wondering about if there is a factory reset option. Also wondering if I've damage the VFD. It is a Moeller DV51.

Any ideas?

Thanks

p.s. Start making up strain relief arms for your spindle cable - you have been warned!

[IanParkin]

Cant help you with the inverter other than read the manual for any ideas but as to the the plug on the top of the spindle I removed the end cap and removed the socket and then drilled for a cable gland and hard wired the wires in with a good earth connection with 1.5mm2 cable to the inverter ..
The plug and socket doesn't look good to be carrying 5-8 amps at 240v
Ian

[IanParkin]

Have you done whats mentioned on page 65 of the manual?
Have you rewired from inverter to spindle?

[routercnc]

Hi Ian,

Yes, I didn't like the look of the connector from day 1 but it has worked ok up until now. If I had bothered to make the strain relief arm I think it probably would have lasted.

On page 65 of the manual I'm looking at it just talks about setting parameters:
"You can adapt the DF51 to your specific applications. To do this,
you need to change the frequency inverter’s parameters with the
built-in keypad, the optional keypads DEX-KEY-…"

Did you mean this or is there another manual you are looking at?

Also, to be clear, the spindle is not connected to the VFD, and the cable to the spindle is completely removed, so wiring the spindle direct won't help at this point. The VFD won't get past the error code on startup. I wonder if something is damaged inside.

I've managed to find all the settings for the VFD which I made 4 years ago, so if I can get it to work with after a reset I can get it going again quite quickly. Just can't find out how to do it . . .

[longy]

Does your spindle have the GX16-3 connector ? Have a look at the PLUG as well and see if there are any burn marks on the female pins they are split as well. I t could be the cable from the VFD has a break somewhere and that has shorted back to the spindle.

E14 "The inverter is protected by the detection ofground faults between the inverter output and themotor during powerup tests. This feature protectsthe inverter, and does not protect humans."

Sounds likely that there is a break in the cable which has flashed at the GX16 connection causing the split in the pin to draw together.

[JAZZCNC]

Been there done this and it cost me VFD + Spindle.!! . . . . . . I made the strain relief from start but temp removed spindle to check some spindle mounts I'd made for some one as didn't have a spare spindle handy but didnt put it back straight away as I had more to make then it got forgot until it did excatly what yours did.!! . . . . . Hope you haven't fried the spindle as well.!

[IanParkin]

I just searched for your model of inverter and page 65 had the error messages and how to reset (not like you refered to )

[IanParkin]

This is the manual I looked at

http://www.inverter.co.uk/applications/manuals&datasheets/moeller-dv51-manual.pdf

[routercnc]

Hi Ian,

OK yes I've tried that. Press stop, activate the digital input configured as RST (this mean links terminal P24 to terminal 5), and turn off. I've tried this lots of times. Fault is still there on start up.

Note that I don't have the spindle connected (UVW terminals) and it still faults E14 (earthing of UVW), so something must be broken inside the VFD.

Looking at my options now:
Have a look inside
Send off for repair - economical?
Buy a new one - I see that the Chinese 2.2kW VFDs are about £90 on ebay (my VFD was £360!)

Question, can I connect a 2.2kW VFD to my 1.5kW spindle? That would leave a future option of a 2.2kW spindle to give me larger collets.

I think that spindle itself is OK because there is about 2.2ohms between each pair of pins. Not a complete check I realise but promising.

Thanks

[IanParkin]

I have not worried about connecting a bigger VFD to a smaller motor
and indeed on occasions the other way for a test off load

just set the max current appropriate for your motor

I have 12 vfd's driving bits of machinery in my workshops and only 1 chinese branded ( i think they all made in china now)
the european brands are easier to set up perhaps ....manuals seem easier to understand
But they all perform as expected
so in your situation I think I would buy a chinese one for £90 and look into having yours repaired as a back up
or buy a package of spindle and vfd for £220 ish? if you thinking of the 2.2kw one
Ian

[magicniner]

It might be worth a look inside, rubber gloves and extreme care around the caps though!
I've always regarded VFDs as a "Black Box" solution but when I recently had an unexplained bang/flash incident with a 750W Siemens inverter drive which then presented errors I decided it was worth a look.
When I stripped it I found a flexible power link between two internal boards had blown but the motor metered out fine and there were no wiring faults or other visible signs of shorts or damage, it had been running for months without any problem before it blew so I assumed the fused power link was a symptom rather than the fault but in desperation I made & fitted a new link and it's been working fine ever since,
Regards,
Nick

[george uk]

I know it may seem obvious.

But i would not recommend opening the VDF for a good 30min after it has been powered to allow any caps to discharge.... apartr from that, it should not be to hard to identify blown componants, or indeed, vaporised earth connections. If i can not spot blown componats, i usually smell them or listen carefully when switched on. :pig:

[routercnc]

Thanks for all the comments and advice. I'm away from the workshop for about a week so time to think about next steps.
Yes, plug is a GX16-3 and 2 of the female pins were worn away at the split line but not burnt as far as i could see. I suspect the continuous movement had worn them and eventually caused a spark.
I'll also check the cable when I get back and look inside the vfd

Starting to research new vfd and spindle. Any views on whether 3kW spindle is worth going for over the 2.2 to give more torque at low speed?

[Jonathan]

It keeps telling me there is a fault (E14 - UVW to earth fault).

Given that you've disconnected the spindle, reset the VFD and the fault persists, that implies there actually is a connection between one or more of the UVW terminals to earth. The simplest explanation for that would be an IGBT that has failed to make a short circuit. I'd have a look inside as you should be able to identify this with a multimeter. If it uses an IGBT module then that will be quite expensive to replace and probably not worth the risk, but if it's individual IGBTs it's probably worth a shot - the issue is you don't know if any of the associated circuits (e.g. gate drive) have been damaged.

No problem getting a higher rated VFD than you need, within reason.

If you need the extra torque from the 3kW spindle then clearly get it...


Been there done this and it cost me VFD + Spindle.!!

You should try rewinding it - shouldn't be difficult.

[EddyCurrent]

Agree with Jonathan's post, here's a supplier I find very good Inverter Drive Supermarket - Variable Speed Drives for Electric Motor Speed Control (http://www.inverterdrive.com/)

[routercnc]

Ok, I've had a look inside and nothing is visibly burnt or smalls burnt (although it was over week ago that the fault occured).

I checked continuity between UVW terminals and there is no connection between them. I didn't have time to do any more investigation but have today made contact with a VFD sales and repair company. They are offering a free quote if I can arrange postage to them, which is fair enough.

Don't know how much the repair will be yet. If it is less then £100 then I'll just repair it. If it around £100 (similar to a new Huanyang VFD), I'll have to think about which way to go. If it is £150 or over then it will most likely be the Chinese VFD and fingers crossed.

I know the risks with these units and the most popular 2.2kW unit seems to be the HY02D223B. I also noticed that they make a SL-222EE model which shows Vector Control Drive as a feature, and has also been shown with an optional removable front keypad panel and connecting cable. This panel can also have a pot added to it to manually control the speed. This would make it the same as my Moeller drive for which I was about to mount the keypad in the control box door. Anyone bought and used one of these models with the Chinese spindles?
I have found a long post from kajuk:

http://www.mycncuk.com/forums/routers-spindles-controllers/7085-need-help-sensorless-vector-vfd-2-2kw-motor-3.html

but he was having alot of set up problems despite Eddy's extensive suggestions for help.

As far as I can tell the spindle is OK. Checking resistance between pairs of pins shows about 2.2ohms. Does this seem about right? I know this is no guarantee of not having a short under actual load, but I'm willing to keep it for now and just fix or replace the VFD.

Finally I need to make sure that the cable does not spark again. I like Ian's idea of going direct so am looking into a gland/strain relief unit like this:

Spiral Cable Gland Strain Relief 4-8mm M16 Thread | eBay (http://www.ebay.co.uk/itm/291127273408?_trksid=p2055119.m1438.l2649&ssPageName=STRK%3AMEBIDX%3AIT#ht_1738wt_940)

I would remove the GX16 socket from the spindle end cap and wire the VFD cable direct to the motor, with an earth tapping made on the inside of the end cap. Might need to open out the GX16 socket hole, or better still tap it to suit the gland thread, but that should be no problem. What other neat cable solutions have others tried? Or are you still using the dodgy GX16-3!?

[Jonathan]

Check the resistance between each of the UVW terminals and the DC bus, as if just one is short circuited you'll still measure a high resistance between the phase connections.

I wouldn't worry so much about the Chinese VFDs - how many recent threads have you seen where the VFD hasn't worked, due to the product not the user? Also adding a potentiometer to either isn't difficult, so I wouldn't make a decision based on that.

2.2 ohms sounds about right for the DC resistance.

Edit: Also, regarding whether to get vector controlled or not - with vector control you will get more torque from the motor (at mid-low speed) and better efficiency, but if you're currently happy with the torque from the motor then it's not really gaining anything so you might as well go for the cheaper option.

[routercnc]

Thank you Jonathan. Here are some pictures for reference:
12498
12499
12500
12501
12502

Are the IGBTs the 3 pin components with a small heat sink? I was vaguely expecting 4 pins but am no expert.

To get to the DC bus I guess I need to get to one of the big capacitor pins? There are 2 boards stacked on top of each other with the top one looking like the high power side and the bottom one looking like the control/software side.

It looks like quite a job to take the top board off to get to the pins since they've soldered multiple pins in place around the edge of the boards to hold it in place (or prevent tampering !).

I'll have another look, but most likely going to send it away.

Found this article quite useful . . . .
The ABCs (and 1-2-3s) of variable frequency drives | Motors/Drives content from Machine Design (http://machinedesign.com/motorsdrives/abcs-and-1-2-3s-variable-frequency-drives)

[Jonathan]

The IGBTs will either be 3-pin components (6 of them - like in the picture you linked to), or a single module containing all 6. Either way they'll attached to the main heatsink as they are the components that switch the power through the motor, so have the highest losses in the driver. I don't think you can see them in your images, as they're probably on the lower PCB.

Yes - the capacitors are connected to the DC bus so if you can get to both pins of a capacitor, that's the same thing. They seem to have missed out one capacitor on the top PCB (the big circle), so you can connect your meter to those pads. The capacitors say 400V on them, so they will all be in parallel.

The 3-pin component in the 2nd picture is some sort of MOSFET to make a switching power supply - it's probably connected to the yellow inductor to derive the logic supply from the DC-bus.

[routercnc]

Had no time yet again this weekend to look at this, plus the VFD sales and repair company seems to have lost interest.

So, finally decided on a new Huangyang 2.2kW VFD (HY02D223B). There are lots of these on ebay shipped from Portsmouth at £100.07 inc delivery etc. So has anyone had any dealings with these companies for example, especially returning any faulty ones or other good service:
hotitemsale
industry-village
qiandingzhensatisfaction
industrysky

Or any other Portsmouth import trader suggestions?

thanks

[routercnc]

Ordered VFD from qiandingzhensatisfaction yesterday. Should be here Monday . . .

On a related note. I'm trying to remove the end cap from the top of the spindle so I can hard wire it and a get a good earth. The strain relief gland has arrived and it looks like it will be perfect.
So far I've removed the 4 cap heads but it is stuck fast. I've tried holding the main body in a vice (soft jaws) and using the largest pipe wrench I've got tried to turn and pull the end off but no luck.

All the photos and info I've gathered over the years on end cap removal only shows removal of the 4 cap heads so it must be a slightly press fit and/or a bit of corrosion. Before I start getting more aggressive with it I thought I'd check if I had overlooked something or if others had found the same?

Thanks

[Clive S]

Ordered VFD from qiandingzhensatisfaction yesterday. Should be here Monday . . .

On a related note. I'm trying to remove the end cap from the top of the spindle so I can hard wire it and a get a good earth. The strain relief gland has arrived and it looks like it will be perfect.
So far I've removed the 4 cap heads but it is stuck fast. I've tried holding the main body in a vice (soft jaws) and using the largest pipe wrench I've got tried to turn and pull the end off but no luck.

All the photos and info I've gathered over the years on end cap removal only shows removal of the 4 cap heads so it must be a slightly press fit and/or a bit of corrosion. Before I start getting more aggressive with it I thought I'd check if I had overlooked something or if others had found the same?

ThanksWhen I took mine off (2.2Kw Chinese spindle) I just removed the bolt and pulled it off I believe some might have a bit of silicone as well.

I think Jonathan has put up a breakdown of the spindle. ..Clive

[Jonathan]

All the photos and info I've gathered over the years on end cap removal only shows removal of the 4 cap heads so it must be a slightly press fit and/or a bit of corrosion. Before I start getting more aggressive with it I thought I'd check if I had overlooked something or if others had found the same?

With those bolts undone it will come off - just a bit of 'persuasion' needed.

[routercnc]

Thanks guys. 'Persuasion' it is then . . . .

[IanParkin]

Just give it a sideways knock with a soft/copper hammer

[routercnc]

Had a few hours free in the garage this afternoon and got working on the spindle wiring modifications.

End cap came off in the end with more of a sideways pull than the pull-and-twist method I had been using before (thanks for the hints!). There was a trace of sealant around the o-rings which must have been holding it on. Don't see why they used o-rings and sealant. The o-rings are recessed into a shallow pocket and look like they would make a good seal once compressed.

Cleaned all the mating surfaces. Removed the GX16 plug and binned it. Dropped the strain relief straight in (mounting hole of GX16 is 16mm, and strain relief also needs a 16mm mounting hole) and used loctite on the inner nut to make sure it didn't vibrate loose during machining. Pushed the VFD cable through and soldered it onto the motor cables. I used 2 layers of heat shrink over each joint just to be sure.

I had noticed that the end cap was painted black on the underside as well so my plan of using that as an earth point wasn't going to work. Luckily on the spindle body side there is a plastic cover cap held in with three machine screws. Checked these and they were a good earth to the outer body of the spindle, so crimped on an earth terminal and fastened it under the screw head.

All wired up:
12546

I will purchase some new o-rings just in case these have taken on a permanent set - if they decided to leak at some point I wouldn't know until something went bang. For now I've re-used them just to test that it works electrically.

Screwed it together and tighten up the cable gland:
12548


New Huangyang 2.2kW VFD arrived Friday (quick!). It is visually slightly different to the photos on ebay.

I offered up the old spade connectors from the mains line filter and they are much too big for the terminals. The small red colour crimp terminals would fit but then the wire doesn't fit. So I ended up filing the sides down:
12547

All for now.

[routercnc]

Last night I wired up the invertor as per this diagram.
12553

R = live
S = (not connected)
T = neutral
9 = earth

Note that I did not connect the spindle motor at this stage (UVW) since I wanted to check it would turn on, and then I was going to set up all the parameters.

Anyway, after double and triple checking all connections I stood back and switched in on at the mains. There was a brief pause, followed by a loud bang and a flash. Great.

I emailed 'qiandingzhensatisfaction', and he replied by sending out the connection diagram above (which I already had!), and asked me to make sure I followed it. I've replied that I had definitely followed the connection instructions and await his reply.

Could I damage the invertor in this catastrophic way simply by not connecting the spindle and switching it on at the wall socket?

Luckily as the unit had just arrived I've not left feedback yet, plus I purchased it on my credit card so should have consumer protection if it goes that far.

Saga continues . . .

[GEOFFREY]

Sounds like a bit of bad luck.It does seem you are covered, but I know just how annoying it is when waiting for something and then it arrives faulty. Await the next chapter with interest!!! G.

[george uk]

Hi

did you use new wiring from the power to the VDF, or did you re-use the older wiring, the boot back from when the first one blew could have affected the internal shielding of the wires,

[routercnc]

Hi George,

Yes I did re-use the wiring from the previous invertor. But I don't think it is a problem because last night I swapped the mains cable back to the original Moeller unit and it came on as before and showed the fault code again (no bang!). I also checked the plug fuse and that is OK. But just to double check I'll do a resistance check on the mains wiring to rule that out.

Given that all I've done is connect three wires EXACTLY as per the instructions, and it went bang on the first switch on, it seems a pretty straightforward case but lets see if the supplier shares that view. Out of interest the email name contains 'lovehappyshopping' in the title so it supports the theory of a few suppliers using multiple outlets under different names.

Thanks

[Jonathan]

That's really unlucky... Accidentally connecting the mains to UVW instead of RST makes it go bang, but only when you press 'Run' and you've not done that.

If they don't replace it then I'll sell you one of my VFDs for a reasonable price as now I can make my own (http://www.mycncuk.com/forums/threads/7576-Build-Log-Servo-Drive-and-VFD-with-vector-control) (potentially) better ones, there's not much point me keeping them.

[george uk]

Hi

I dont know enough about the operations and standing loads of VDF to make any useful comments rc, i just added the bits about the wiring to it as i have made majic smoke with electronics a few times before... and have noticed after strip down from large pops, degradation of insulating wires around the dump or earth load, there is one that springs to mind with HID lighting, it seemed fine unter test, but kept blowing ballasts, after stripdown of the wires, the wires had dried and slightly swelled. Under low load test, things seemed fine, but would blow windings under load, especialy once it had dried out the cable enough..... there was no sign on the outside of the cable of the swelling/drying

Its sounds more likeley that its a dodgy component and you have just been very unlucky,

[routercnc]

My replacement VFD arrived last night. But they had sent me the 1.5kW version despite me originally ordering the 2.2kW unit.

They obviously weren't happy with me connecting to a 1.5kW spindle.

Fed up with it all I decide to wire it up and see if it works. Pleased to
report that after a few seconds delay it fired up. I programmed in all the relevant parameters. If I get the chance this weekend I'll connect the spindle and check it all works.

One thing I noticed was the noisy fan. On the Moeller unit there was a setting to allow it to only come on if it reached a certain temperature. Couldn't see anything in a quick flick through the manual but not a big deal if it does not do that.
Also the keypad feels very cheap! Luckily it is removeable and has a connector in the back so would be seconds to replace if they are available as spares.

[JAZZCNC]

One thing I noticed was the noisy fan. On the Moeller unit there was a setting to allow it to only come on if it reached a certain temperature. Couldn't see anything in a quick flick through the manual but not a big deal if it does not do that.
Also the keypad feels very cheap! Luckily it is removeable and has a connector in the back so would be seconds to replace if they are available as spares.

No Can't do anything about the fan it needs it on all the time. Regards the Keypad then yes that's a known fail area over time with these VFD's. Thou if you use spindle speed control thru Software then you hardly ever touch the Keypad so it's not an issues.
If you don't Have spindle speed control then Fit a potentiometer to it and again you'll hardly need to touch it other than pushing run & stop. Most of the keypad use and wear comes from playing and setting speeds. Some where on Forum Clive S described how to do it and what to buy. Clives a Good old lad and he'll be along in a minute to tell you where to find it or how to do it.!

[Jonathan]

You should ask them to refund the difference as the 1.5kW VFD is cheaper. They probably just sell them and don't know much about it, as there's nothing wrong with using the 2.2kW VFD with a 1.5kW spindle. Either that or they're trying to cut their losses.

The few seconds delay is normal - it's just to reduce the inrush current when charging the capacitors.

It's very simple to just put a temperature sensor on the heatsink and switch the fan on/off accordingly. You'll probably find it doesn't need to be on that much, since most of the time the spindle (and therefore VFD) won't be operated anywhere near rated current.

[Clive S]

This might help with the settings for the VFD:- http://www.mycncuk.com/threads/7388-Fitting-a-pot-to-a-Chinese-Spindle-VFD post 1 .. Clive

[routercnc]

Hi Jazz,
OK, thanks for that. I can make the fan quieter with a bit of acoustic treatment.

Clive,
Thanks for the info. Start/stop and pot has worked well for many years on the old VFD so will most likely go with that option.

Jonathan,
Yes, first thought that crossed my mind was that they owe me about £15 or so! I'll email them and see what they say.

I've settled on getting this one to work - if I go down the route of sending it back an insisting on the 2.2kW unit, and it is faulty (or develops a fault), it could be a difficult discussion.

This afternoon I managed to connect up the spindle and when I pressed run it smoothly ramped up to 24000rpm. It ran in the wrong direction, but easy to fix that one. Another step closer . . .

[JAZZCNC]

Hi Jazz,
OK, thanks for that. I can make the fan quieter with a bit of acoustic treatment.

Why bother can't hear the bloody thing when machines cutting and when not using turn it off.?

Personally I would be getting onto them and insisting on the 2.2Kw just for the fact it gives larger cutter choice and if replacement was faulty then I'd be asking for my money refunded and buying else where. You have done nothing wrong running 1.5Kw with a 2.2Kw VFD provided you used the correct parameters for the spindle and infact your running the VFD well within it's capabiltys so better for them as it's less likely to be stressed.!