IGaging DRO EMI Problem

[kcoffield]

I'm going to make one more run at fixing the RFI/EMI problem because I suspect I may have other issues in the shop induced by the VFD later down the line and I may be able learn something here and maybe it will help someone else. I have a CNC router with long run times and soon to be 3D printer with same. I've already had the CNC router exhibit self induced RFI problems that needed to be fixed. Wouldn't be the end of the world if I could run the lathe while those were running but certainly suboptimal.

There is definitely a conducted EMI problem with the VFD too because with it switched on, I can't power the iGaging DROs from the wall wart in any shop receptacle away from the lathe, but switch off the VFD and it powers up and displays well from anywhere, so the VFD is conducting back EMI/RFI through all my shop circuits.

I bought some cheap parts but they may take a few days to receive them with the holiday season. I do have couple questions I'll pose later below.

If no success, I'm happy to just run the DROs off of battery but it won't be that fancy battery pack in the post above. Turns out, that battery is meant to re-charge a phone battery not be a power source, and when it senses low draw for 30 seconds or so, it shuts off as sensing it has charged the phone battery. The DROs don't draw enough to keep it active. So some of what I thought were the DROs being taken down by RFI while the battery was plugged in was really just the battery pack shutting off, and when that happens, the DRO bypasses the internal coin cell battery because it senses the barrel plug in the socket, but doesn't have power......good grief.

One other tidbit I hadn't previously mentioned, when I rec'd the VFD, as I was reading the manual, it said it cannot be operated on a GFI circuit. Well, code required all the 120vac circuits in a garage to be GFI. There is a custom version of the VFD for GFI circuits that has both hardware and software differences and cost twice as much so probably wouldn't have bought it even if I had known. I just removed the GFI receptacle and installed a non-GFI receptacle. I know what the GFI does but not sure how/if the mechanism could affect this conducted EMI problem. I think the reason the VFD won't work on GFI circuit has to do how it senses current for overload protection and maybe even switching functions. This may also prevent me from installing any inline suppression devices which may mean the lathe simply doesn't run while other sensitive equipment in the shop runs. I'll put that question to the VFD manufacture.

So the RFI fix questions are:

1. I have braided shielding I can pull over the cables. Probably won't do anything but I have plenty of it. Is suspect it is better to shield the power cables since they are the source rather than the control cables. Is that so? Could do both and I will reroute the control and power cables apart from one another to extent possible.
2. I bought an assortment of ferrite beads, two dozen for $9 delivered in different sizes. I have a few spots on my CNC router that could use them too. They are cheapies and I don't know what the core material is or intended frequency range, but even if I did, I wouldn't know which one to select. I'll try them in single pass and looped. Should they be placed on both ends of the power cables? I may also place a couple more on the supply power to/from the VFD. I presume the cable should not be have braided shielding in the area of the beads. Is that so?
3. For the conducted part of the problem, I bought a 12vdc to 5vdc Buck Convertor, thinking my 12vdc from the regulated supply seems to be a clean source. It's only a $10 part but I need one for my ATV so if it doesn't work on the DRO, I'll use it there. I don't know much about those. Is it like to be another source of noise?

Funny how everything becomes a ball rolling down hill. Thanks for playing along.

Best,
Kelly

 

[chip maker]

No worrys. The short answer is because I already had to run the control cables, had the wall wart, and the power cable extensions were only $4......so why not.......answer: because they introduce EMI/RFI. So why did iGaging provision the DRO with a power port?

If you only have 5 DROs total on two machines, and you don't go through hardly any batteries, why to you buy them by the 100s?

Best,
Kelly

 

[chip maker]

Sorry misprinted I purchase them by 25 at .36 cents a piece. I not only have the 5 Igage but also about 4 other things that use the same battery size. At the regular stores these are like 3.00 for 2 batteries.

 

[markba633csi]

I think you will find the ferrite cores most effective right close to the dro units, as close as possible. So run the shielding up to the cores but not thru. Use one core for the power cables and one for the controls
Now, as far as grounding the shields, that might take some experimentation and you may not need to ground them at all. Can't advise you there.
As far as the buck convertor, it's worth a shot.
-Mark

 

[kcoffield]

I received the Buck DC-DC convertor and installed it downstream of my 12vdc voltage supply.



Like before, fine with the VFD off but no display on the DRO with the VFD on, and the display came right back when I toggled off the VFD.

So then I installed ferrite bead with 3 loops on the DRO end of the power cables.



Still the same result when I added a bead on each end of the cables…..scrambled and or blank display when the VFD was switched on but recovery when it was switched off.

The cables were separated from each other in all instances. With those results I wasn’t willing to put the effort into pulling braided shielding over the cables. It doesn’t seem to have worked for anyone who has posted, but I did decide to see if isolating the scales from ground had any affect.

The easiest way for me to test that was to unmount the cross slide DRO, so I did…..It had no effect.

I didn’t try adding a cap across Vcc and ground, even though it appears that is the only thing that has worked for Homebrewed and RJSakowski in the other thread.…..

Battery replacement for battery-only Igaging DROs?

RJ, what size was the cap and can you elaborate a little on exactly how/where it was installed? I have a similar problem described here: https://www.hobby-machinist.com/threads/igaging-dro-emi-problem.89275/ Best, Kelly I used a .05 mfd ceramic disk capacitor in parallel w2ith aa 10 mfd...

www.hobby-machinist.com

…..it’s just not worth at this point to me for what might be a more or less a non-issue. If the extensions of the control signal cables didn’t work that would have been a bigger problem for me but running the DROs on batteries is not.

I will say, I cannot plug the DROs into any kind of remote power source in any circuit in my shop while the VFD is switched on which makes me think that it is really polluting my power source and I suspect I won’t be able to operate the lathe along with any other sensitive equipment in my shop. Either KB’s VFD is a very bad actor as far as back EMI or iGaging did zip to protect there displays when they added the remote power provision. I may eventually need to place a call to KB on this.

Best,
Kelly

 

[pontiac428]

@kcoffield, you should try the iron beads on the signal wires, that's what they are meant for. Your power wires should not be an issue.

 

[homebrewed]

I think you will want to connect the shields to each DRO's internal ground bus, IF you decide to try it. When I was fighting my EMI problem I put shields over the cables and found that the best EMI reduction (not elimination unfortunately) was with that connection. Problem is, the ground bus isn't readily accessible unless you either: (1) cut open the USB cable and access the ground line; or (2) open the display box, solder a wire to GND and bring it out of the box. Both make for an ugly look, to put it mildly. In your case, GND is sort of available via the external power receptacle, but if you plug anything into it, the internal batteries are disconnected. Sort of a conundrum there.

I have a suggestion that might work for you, considering the fact that your DROs work OK on batteries. It will require a modification of the external power cable, and getting a few capacitors and resistors. There will be a small voltage drop across the resistors, but the current generation of iGaging DROs draw so little current that it will be just a few milli-volts. Anyway, here's the circuit:



To make this, cut the power cable close to the connector and expose the wires on both sides of the cut. You will want to make sure that the wall-wart's +V is connected to the plug +v (through R1), and the same for the GND connection through R2. Reversing the connections will likely kill your DRO, you have been warned! Use heat shrink tubing to insulate the connections, then a larger-diameter piece to cover the whole thing up. You can use an electrolytic capacitor, just make sure the "+" terminal connects to +V if you do. Try this on one DRO first to make sure it WILL work, eh? And double-check that the polarity is correct on the plug terminals before you plug it into your DRO. The wall-wart should indicate if the center is positive or negative.

It is a filter that will remove the EMI and keep it from polluting the DRO's Vcc and GND busses. It actually improves on what I did to fix my problem, but without needing to open up the sensor box and cut a circuit board trace to do it.

I wouldn't use this in combination with shielding. Connecting shielding up to the DRO ground would likely reduce the effectiveness of the filter network. If you want to use shielding, connect it to the wall-wart GND line.

 

[kcoffield]

@kcoffield, you should try the iron beads on the signal wires, that's what they are meant for. Your power wires should not be an issue.

Since I still had the buck convertor in series with the 12vdc power supply, it just took a couple minutes to wrap on the bead and try.....but same result, no operation with the VFD switched on and power supplied instead of battery operation.



They always scramble and then completely shut down. When the VFD os off or revert to battery power they recover, but always in mm units of display.

....I have a suggestion that might work for you,....

I can do that, but not being an electronics guy, I don't have those components or anything to harvest them from so will need to buy them, but looks like I can do so and get a jillion of them delivered to my door for under $9. I seem to have a good collection of low quality power cables to experiment with so that shouldn't be a problem. Anything special about the resistor as far as power or type of cap?

Best,
Kelly

 

[homebrewed]

Anything special about the resistor as far as power or type of cap?

The current draw is extremely low, so quarter-watt or smaller resistors would be just fine. Just about any capacitor in the low-microfarad range will work, whether it's a film capacitor, ceramic capacitor or electrolytic. Since the values aren't all that critical you can go with 10% tolerance values on the resistors. Capacitor tolerance can be +/-20%.

I don't know who your go-to vendor is for electronic components, but just for examples on what would work I went to Jameco.com and did a little searching. For capacitors, something like their P/N 10866 would be fine. It's a 1uF 50V 20% electrolytic that costs $.59 in singles (so you will need to connect them up correctly). The best price I found for a 1uF film capacitor was $.79 (P/N 231239). A .25 watt 1K 5% carbon film resistor will cost you $.06 but you have to buy ten of them, oh dear . P/N 690865. The shipping may cost more than the parts.

BTW, the sensor is powered from the USB cable (what you're calling the control cable). Since the display unit provides the power that means there's a good chance that the power at the sensor end is polluted with EMI, too. RJ and I both found that installing capacitors between Vcc and Gnd at the sensor end helped a lot, so you might as well order enough capacitors to do that if you need to.

 

[kcoffield]

.....I don't know who your go-to vendor is for electronic components, but just for examples on what would work I went to Jameco.com and did a little searching.

Not being an electronics guy, I don't really have a go-to, and thanks for that lead. Turns out Jameco has a $10 surcharge on orders <$20 so that would be on top of product and shipping cost. They're common parts so being a Prime member shipping is free, so for ~$9 I have 100 resistors and 50 caps heading my way via Amazon......says Tuesday delivery but in any case should be next week. I'll let you know how I do. Thanks for your help.

Best,
Kelly