2

I don't know if it is right place to ask, if not correct/guide me.

I recently have built treadmill workstation, where I could walk and work (rather than just sitting).

But as soon as I turn on treadmill, nearby keyboard, mouse and monitor starts glitching:

  • Monitor turning off/on by itself (occasionally)
  • Keyboard repeats keypress (press "s" once but on textfield you see "sss") (most of the time)
  • Keyboard turning off/on by itself (frequently)
  • Mouse just turns off until I reconnect USB (rarely)

Which makes my experience awful and I had to switch back to sitting.

Treadmill itself works perfectly, I can't complain or explain problem to seller.

What I have tried:

  • Connected EMI filters to PC

    I thought treadmill might be somehow modifying AC sine wave since it can act like generator or generating noise.

  • Wrapped underside of the treadmill desk with conductive fabric

    I thought glitches caused by magnetic/spikes that treadmill is generating which might be adding more noise into HDMI/USB cables.

    p.s. I tested conductive fabric with my phone, it blocks WiFi, cellular network and bluetooth perfectly.

After wrapping with conductive fabric what I have got:

  • Monitor turning off/on by itself (occasionally)
  • Keyboard repeats keypress (press "s" once but on textfield you see "sss") (occasionally)
  • Keyboard turning off/on by itself (rarely)
  • Mouse just turns off until I reconnect USB (rarely)

What do you think, what causes such glitches?

JavaMachine
  • 131
  • 7
  • 1
    One test you can consider, expensive though, is to try and protect your computer equipment with a large isolation transformer. It is very hard for glitches to get through those. Maybe borrow one? You might also try a UPS. – jonk Feb 21 '19 at 12:21
  • 1
    @jonk - While I have been preparing my answer you have nicely summarised two of my key suggestions. Nicely done. No, I wasn't peeking. I have a broken shoulder (clavicle) so am using google docs to voice dictate and pasting the edited text. An OK solution. – Russell McMahon Feb 21 '19 at 12:47
  • 2
    @RussellMcMahon I was just tossing out two things from experience, quickly. Had to deal with a dental X-Ray machine in an office space below ours, once. The glitches were ugly when they used it. Didn't have UPS, back then. So used a big isolation transformer to cure it. Treadmill motor drivers, those I've seen, are too simple to soak up human energies and they don't often include large multi stage emi filters on the ac in side (big beasts, and expensive, done well -- so they don't do it ) – jonk Feb 21 '19 at 12:56
  • 1
    @RussellMcMahon Sorry to hear about the broken shoulder and the difficulties that means for using a computer, among everything else that means. That sounds like it will take some serious time getting back to baseline. – jonk Feb 21 '19 at 13:02
  • 3
    You have made quite clear that the noise comes from the treadmill, but your solution is to shield the PC and its cables. Even if you were successful at that, it is likely that something else in your/neighbours house will still suffer from the noise. The most sensible approach would be to remove/reduce the noise as much as possible at the source. Add filters as close as possible to the source of the noise. Is this a commercial treadmill or did you make it yourself? – Elmesito Feb 21 '19 at 13:07
  • @Elmesito commercial treadmill, LifeTop LT3440 – JavaMachine Feb 21 '19 at 13:08
  • 1
    Questions regarding the use of consumer electronics are [off-topic](https://electronics.stackexchange.com/help/on-topic). – brhans Feb 21 '19 at 14:13
  • 1
    @brhans No. That statement is incorrect without qualification. The guidelines which you are (presumably) citing explicitly state (non verbatim) that questions re attempts at modification , repair or remediation are valid if they involve electrical action by the OP, a degree of effort and understanding etc. I'm up briefly from bed, and am somewhat immobilised with a broken shoulder. Please provide a ref to guidelines you and thus also I am citing so we can both learn. – Russell McMahon Feb 22 '19 at 15:42
  • @RussellMcMahon - The topic policy for this site is right there in the link in my comment, where it's always been.You are also surely aware of the standard Vote-to-Close reason which states "Questions on the repair of consumer electronics, appliances, or other devices must involve specific troubleshooting steps **and demonstrate a good understanding of the underlying design of the device being repaired**." and this question falls far short of that threshold. – brhans Feb 22 '19 at 22:25
  • 1
    @brhans Link to ref, as noted, would be useful. "You will also surely be aware" that you are quoting back to me verbatim the guidelines that I said I was quoting non verbatim. | If you can read the detail in the question, note that this is a newcomer and that the "even nicer than usual policy" applies, and STILL not see that he has described 'specific trouble shooting steps', and tried much harder than most to act logically and to lay out what he has done and why he did it and what he hoped for and what happened (or didn't) then you have failed my course, please take your knives and leave :-). – Russell McMahon Feb 22 '19 at 23:47
  • 1
    @RussellMcMahon - The link, as already noted, is there - in my comment, should you choose to click on it. I fail to see the problem in quoting this sites policies verbatim - they're not religious texts which require interpretation. The fact that the OP has tried to write a good question does not make that question [on-topic here](https://electronics.stackexchange.com/help/on-topic). A vote to close with a link to the reason hardly constitutes "not being nice" when *that's how Stack Exchange works*. Perhaps a refresher of the [tour](https://electronics.stackexchange.com/tour) would be in order? – brhans Feb 23 '19 at 02:07

2 Answers2

4

You have EMI issues, as you have deduced.
This can be radiated or conducted via wiring.
The problem may be poor EMI suppression in the treadmill, or poor PC system EI rejection or, quite possibly, some mix.

Without capable test equipment it is difficult to quantify the magnitude and bands of noise transmission and the location and types of noise sensitivity BUT, lets just fix it instead :-).

If you can operate TM (treadmill) and PC (PC system components) from two independent power supply phases it MAY help. But, only maybe. If operating from a single mains outlet.
Use two "plug boxes".

Throughout the following tests I repeatedly say “report the results”.
the intention is to identify changes in noise level which may be made by changes in configuration as various combinations are tested. Failure to record results will lead to uncertainty as to what did and didn't work.

Operate all PC equipment monitor etc with mains power cords going directly from a single plug box to the equipment.
Operate the treadmill from a separate plug box.
Report the results.

You mentioned having an EMI filter.
Plug the treadmill blood box into the wall outlet and then plug the PC plug box into the treadmill plug box via the EMI filter or, if available, directly into a separate wall socket.
If you have a second EMI filter that is capable of handling the treadmill peak current use that to connect between the treadmill and the wall outlet.
Try operating the system as described and report the results.

Next if you are able to borrow a UPS (uninterruptible power supply) capable of running the PC or the whole PC system, operate the PC via the UPS and report the results.

Next, pull the UPS power supply plug out from the mains so the PC system is running on the UPS and report that results.
You would hope that if the signal is conducted via the mains, then running the PCfrom a UPS alone will produce a superior result.

Next without using the UPS borrow an isolating transformer (110 to 110 or 230 to 230 volts AC depending on your local power supply voltage) and with enough power capacity to run the PC or the whole PC system via this transformer and report the results.

Also try running the EMI filter in series with the supply to the Transformer.

Obtain a multimeter with an ohms range set it to low-ohms and with the equipment all turned off but plugged in to the plug box. pull the power plug from the plug box out of the wall socket and Measure the resistance between the plug box ground pin and earthed portions of each Item of equipment, to ensure that there is a low ohms connection between for instance the PC ground and printer ground or PC and keyboard metal and so on. It may be that the PC metal is not grounded. If any of equipment is not grounded make it so. if uncertain about grounding ungrounded equipment seek local professional help. Report the result.

At this stage report back with individual and overall results and we will see if you are getting somewhere or if a different approach seems warranted

Russell McMahon
  • 147,325
  • 18
  • 210
  • 386
0

Same problem here, i just mounted an old legacy touchscreen above an oldschool loud big treadmill. First time ive seen a screen flicker in that manner also first time encountering EMI firsthand...but defo EMI in my case atleast. I tried different power bars with fuses and then different power circuits in the house to no avail! Then just for kicks, i grabbed the braided long HDMI carrying the signal from across the room and lifted it well above the treadmill and that did the job. Still flickers every so often but totally usable/viewable. Just went on for 15 mins- flickered once! Gonna look for some cable sheathing or a piece of emi shielding. It also might be worth mentioning aim also using an hdmi to dvi adapter which may also increase exposure area for EMI interference..

Hope it helps somebody

R-Tech Solutions
  • 9
  • 2
  • 1
    A conductive brush that spans the width of the belt and sweeps it as it runs connected to the treadmill frame should also help. I'm fairly certain such problems are from the static that accumulates then discharges. My treadmill was like a van de graff generator. – DKNguyen Jan 25 '22 at 19:59