pgershon Posted October 2, 2020 Author Posted October 2, 2020 3 hours ago, epfunke said: Any chance of a quick sketch showing the layout. with approximate distances? like Main house:------ 100 ft ----- :pool house panel: 20breaker---10 ft? -----Insteon device :pool house panel: 70a breaker- 15ft? ----- jandy box Some other thoughts... this noise is probably broadbanded with lots of harmonics, you can try a an AM radio tuned as low as it will go and see if you hear any interference when the pump is on vs off? try and use it as a sniffer? Is it the pump motor itself? or the controller? (my bet at this point) this may shed some light. Also if you have a receiver probe for a tone tracer, that may also pick up the noise and help locate it Or is the jandy box some other control and the intelliflow is the pump controller. Knowing more about the topology may shed some light on this Pentair Intelliflo pump --- 4 feet --- connects to 20 amp GFCI breaker in Jandy box (multiple other pumps and pool lights, pool heater, salt chlorinators connect to other breakers in Jandy box too - none are Insteon devices) --- 15 feet --- Jandy box service connects to 70 amp breaker in pool house panel (multiple other breakers in pool house panel including Insteon controlled lights and fan, electric hot water heater, refrigerator, dishwasher) --- 100+ feet ---- connects to 100 amp breaker in main house (lots of other Insteon devices. Further overview - main house has 4 electric panels - generator and solar panels - 100 amp service goes to pool house Will be hard to try AM radio as (1) not sure I have one and (2) limited or no AM reception in area. Jandy box is just a breaker panel. This pump and other pumps connect to Jandy box. Other pumps are single speed, Intelliflo pump is variable speed. Other pumps do not cause interference when they run without Intelliflo pump. Interference ONLY when Intelliflo pump is running - does not matter if running at high speed or low speed. Intelliflo pump has controller built in. There is an RS485 connection to the pump that signals to the controller, but the pump controller can run standalone without the RS485. Disconnecting the RS485 does not remove the interference.
pgershon Posted October 2, 2020 Author Posted October 2, 2020 3 hours ago, epfunke said: That filter may help. one key would be to keep that ground wire as short as possible When you say "keep that ground wire as short as possible", which ground wire do you mean? The one on the filter? I figure I would put filter in panel (Jandy box) and connect ground to the ground in the Jandy box. I was going to run one of the tho loads from the pump as the loan and the neutral on the filter.
pgershon Posted October 2, 2020 Author Posted October 2, 2020 2 hours ago, ELA said: Have you tried the XPNR, as close to the pump as possible, along with the XPF but closer to the pump breaker? The idea being that the XPNR shunts the noise outside of the Insteon band near to the pump and the XPF isolates the Insteon devices from the noise source, at 132Khz and closer to the Insteon devices. Can we get an attenuation specification link to the Uxcell AC 115/250V 20A CW4L2-20A filter? http://www.uxcell.com/115v250v-20a-cw4l220as-power-single-phase-emi-filter-p-349091.html The pump plugs into a 240 outlet and runs directly to the Jandy box. I do not have access to a neutral closer than the Jandy box so the best I can do is connect two XPNRs to the two leads from the pump plug (and the neutral). I could then run the XPF between lean and the breaker - I will try that - its on my list. Very hard to get all this equipment into the Jandy box too - but let me see if it functions first.
larryllix Posted October 2, 2020 Posted October 2, 2020 59 minutes ago, pgershon said: <snipped> Will be hard to try AM radio as (1) not sure I have one and (2) limited or no AM reception in area. <snipped> You aren't looking for stations but rather noise. Tune it between stations. You'll know when you hear it.
epfunke Posted October 2, 2020 Posted October 2, 2020 @pgershon thanks for the description and info about the layout, I now have a better picture of what you have, and @larryllix, you are absolutely correct just tune that am radio between stations, and it will receive the noise, and it may be so strong that will be hard to locate an exact source but is super helpful to diagnose as you turn loads on and off and the noise comes and goes.... but based on what we have I think you are correct it is the intelliflow, and after looking at that pump online looks like the motor speed control is built right into the pump. The reason i talked about the am radio was i did not know if that jandy box was a speed control of some sort but now understand that is all part of the pump itself Good news is we dont have to filter a 70 amp load, just 20amp for that single pump. you already tried the xnf in the jandy box and that did not cure it, so you already own these, I would put those back in the jandy. based on knowing the noise is coming from that pump any way of trying the xpnr filters right at the pump itself? just asking as I dont know if there is room in the junction box on the pump, or just hooking up there as a test? at the pump is there room for a ferrite core? amazon 4ea 38X19X13mm toroid transformer core noise filter ferrite core chokes ferrite bead toroidal ferrite ring loop the vires through the core as many times as possible. is this a new installation, or something that has worked for a while and has degraded? Check that bonding grounds at pump are tight and not corroded? Hope this helps
pgershon Posted October 2, 2020 Author Posted October 2, 2020 1 hour ago, epfunke said: based on knowing the noise is coming from that pump any way of trying the xpnr filters right at the pump itself? just asking as I dont know if there is room in the junction box on the pump, or just hooking up there as a test? There is, in theory, a way to test the XPNR at the pump or the junction box - except that the XPNR needs a neutral and I dont believe there is a neutral before the Jandy breaker box itself. The pump is 220 V so it has two hot leads and a ground. The XPNR needs a hot and a neutral. Am I missing something?
pgershon Posted October 2, 2020 Author Posted October 2, 2020 1 hour ago, epfunke said: at the pump is there room for a ferrite core? amazon38X19X13mm toroid transformer core noise filter loop the vires through the core as many times as possible. is this a new installation, or something that has worked for a while and has degraded? How would I install the core? Which wires would I loop? There is a cable that goes from the Intelliflo to the outlet - I could loop that cord through the beads - that would pass the 2 line wires and the ground through the ferrite core - but it would be exposed to elements - is that OK? It is an old installation and I believe it has degraded over time. Install was in 2012 - I noticed issues in 2018 for first time - worse in 2019 and I figured out the cause in 2020. What is best way to check bonding? I looked yesterday because I was changing the next door spa pump motor - it is connected but looks a bit old and corroded. How best to test it?
epfunke Posted October 2, 2020 Posted October 2, 2020 It is my understanding the xpnr is a two wire shunt type filter, it is connected across the circuit and the xpf is a three wire shunt and low pass that has a neural, and the load current passes thru the black and red leads. the core would be trial and error. the key is when you loop it it must loop in one direction. so if you are putting the entire cord thru the core you may have to remove the plug. key is loops have to be in one direction, if looking at the donunts face and you put the cord through the hole, for the next loop you need to go outside the ring and back through in the same direction as the first time. google ferrite core filters They also make snap on ferrites that are easier but i dont think they work as well but due to simplicity may be worth a try. either way the closer to the pump the better. as far as grounds i would inspect and reattach anything that looks suspect. make sure breakers are off if you disconnect any grounds to clean them as there could be small leakage currents. I would start with that old and corroded looking one.
garybixler Posted October 2, 2020 Posted October 2, 2020 (edited) Having a similar noise problem generated from a single phase to a three phase variable speed controller for a well pump I ended up using this filter before the controller as that is where the noise was being generated. Delta CA302-R 2-Pole 125/250V Single Phase Surge Capacitor. The noise may be a low frequency with harmonics but overwhelms the Insteon traffic. I was able to observe this on a scope. Also the neutral on the pool sub panel may not be grounded. Edited October 2, 2020 by garybixler
pgershon Posted October 2, 2020 Author Posted October 2, 2020 (edited) 56 minutes ago, epfunke said: It is my understanding the xpnr is a two wire shunt type filter, it is connected across the circuit and the xpf is a three wire shunt and low pass that has a neural, and the load current passes thru the black and red leads. The XPNR is two wire. I connected one wire of each to each hot from the pump and the other to neutral at the box. If I connected at the pump, I have two hot leads and a ground - no neutral. I do not believe you want to connect the two hot wires to the sides of the XPNR (or is it OK since one is positive and the other is negative at all times?). Perhaps one to each hot and the other to ground? Here is a link for the XPNR - remember at the pump there are only two hot leads and a ground. http://www.authinx.com/manuals/X10/XPNR.pdf Edited October 2, 2020 by pgershon
epfunke Posted October 2, 2020 Posted October 2, 2020 you can try the xpnr from hot to hot at the pump itself, it says it is rated for 240 vac. Neither hot is positive or negative, they are both ac 60 hz, 180 degrees out of phase, so with a meter you will measure 120 to ground / neutral, and 240 from hot to hot because of the phase difference. Connecting anything between the hots and the safety ground is not a good idea from a safety point of view, and would possibly trip your GFCI anyway you say it was working and then got progressively worse. any other changes in that time?
pgershon Posted October 2, 2020 Author Posted October 2, 2020 No other changes. I thought my Insteon units were failing (some were too, different issue) and it is only this year that I realized the problem is the pool pump causing interference.
pgershon Posted October 2, 2020 Author Posted October 2, 2020 I checked the bonding lug on the pump more carefully. The screw terminal rusted and it looks like a pool man ran the copper wire through a screw hole in the lug but there is no secure contact. I ordered a replacement lug for the pump. May not fix the problem, but a good idea to correct regardless.
pgershon Posted October 2, 2020 Author Posted October 2, 2020 I just tried the Uxcell AC 115/250V 20A CW4L2-20A filter (http://www.uxcell.com/115v250v-20a-cw4l220as-power-single-phase-emi-filter-p-349091.html). Interesting result in that it improved but did not fix my interference. I inserted device between the breaker and the load to each device (with a ground on the line side as well). Used the neutral as a second lead as the comments for the device suggested. The light outside the pool house (Insteon Switchlinc dual band) worked with the pump running - full on/off/dim control. But the fan control (Insteon Fanlinc) could not be controlled from the main house with the Intelliflo running. So it seems the Uxcell AC 115/250V 20A CW4L2-20A gave me partial attenuation of my interference.
epfunke Posted October 2, 2020 Posted October 2, 2020 Great to hear progress. are the Uxcell filters the only ones in circuit right now?
pgershon Posted October 2, 2020 Author Posted October 2, 2020 23 minutes ago, epfunke said: Great to hear progress. are the Uxcell filters the only ones in circuit right now? They were for this exercise. I removed them as it started to rain and I did not want tower with the panel box open. I plan to try the XPNR, with each of the Uxcell and the two XPF filters. Space in the Jandy box is an issue.
pgershon Posted October 2, 2020 Author Posted October 2, 2020 4 hours ago, epfunke said: you can try the xpnr from hot to hot at the pump itself, it says it is rated for 240 vac. Neither hot is positive or negative, they are both ac 60 hz, 180 degrees out of phase, so with a meter you will measure 120 to ground / neutral, and 240 from hot to hot because of the phase difference. Connecting anything between the hots and the safety ground is not a good idea from a safety point of view, and would possibly trip your GFCI anyway I just found the guidance that caused me to install the XPNR in the Jandy box where there is a neutral as opposed to at the pump. What do you think (here is link)? http://kbase.x10.com/wiki/XPNR 240V Applications The XPNR does require a neutral connection and can be used on a 220/240 Volt AC device by putting one XPNR on each of the AC Hot leads.
Goose66 Posted October 2, 2020 Posted October 2, 2020 (edited) 2 hours ago, pgershon said: The XPNR does require a neutral connection and can be used on a 220/240 Volt AC device by putting one XPNR on each of the AC Hot leads. I am not an AC expert (I actually got one of my worse grades in college in Power Systems), but it seems like you will need to attenuate noise between each of the two phases and the neutral - not just between the two phases. Edited October 2, 2020 by Goose66 1
Goose66 Posted October 2, 2020 Posted October 2, 2020 Also, this is another long shot, but I was fiddling around in my pump house the other day and I noticed all of my pumps are grounded through an external lug to a grounding post, this is in addition to the ground wire in the power lines from the breaker panel. Another poster referred to it as a "bonding ground." Does your pump have this as well? I know when I was working the sound board for a theater back in the day we had to connect the "bonding ground" (what we called chasis ground) to a decent ground or we could get some pretty bad 60 Hz hum at high volumes.
pgershon Posted October 2, 2020 Author Posted October 2, 2020 1 hour ago, Goose66 said: Also, this is another long shot, but I was fiddling around in my pump house the other day and I noticed all of my pumps are grounded through an external lug to a grounding post, this is in addition to the ground wire in the power lines from the breaker panel. Another poster referred to it as a "bonding ground." Does your pump have this as well? I know when I was working the sound board for a theater back in the day we had to connect the "bonding ground" (what we called chasis ground) to a decent ground or we could get some pretty bad 60 Hz hum at high volumes. I ordered a new external lug as the pump is not well connected to the copper wire. Not sure if this will help or not, but the pump was not well connected as ois and the connection rusted.
ELA Posted October 3, 2020 Posted October 3, 2020 6 hours ago, pgershon said: I just tried the Uxcell AC 115/250V 20A CW4L2-20A filter (http://www.uxcell.com/115v250v-20a-cw4l220as-power-single-phase-emi-filter-p-349091.html). Interesting result in that it improved but did not fix my interference. So it seems the Uxcell AC 115/250V 20A CW4L2-20A gave me partial attenuation of my interference. I had asked for the attenuation specifications not a sales page. EMI filters with higher attenuation specifications at lower frequencies can be expensive. Some times Home Brew approaches can also work.
pgershon Posted October 8, 2020 Author Posted October 8, 2020 I just went in and (1) re-did the bonding connection on the pump, (2) added 4 ferrite core filters by the pump, (3) re-installed both the XPR and XPNR's (2- XPNR closest to pump, each to a neutral, and the 2 XPR in series). None of this had any effect. Here are photos from my basement switch link in access mode - the fast flashing is due to interference, the normal rhythmic flashing is with pump off. The noise continues to radiate through my house, even with the filters connected (as long as pump is running). IMG_2651.mov IMG_2652.mov
garybixler Posted October 8, 2020 Posted October 8, 2020 This is what I had put together to dramatically reduce the noise coming from my single phase to 3 phase variable speed well pump controller. I couldn't say if it would have the same effect on your setup or not. I was able to see on a scope the ongoing reduction of noise as I added and rearranged components. The CA 302 R's were essential in the noise reduction more so than the FN2410-45-33 which didn't seem to have much of an effect. Also there was no GFCI involved so I wasn't worried about having that trip. My ground and neutral are bonded in the main breaker box 3 feet away. The big problem was that the components are large and I had to put them into a separate electrical box. It might be worth trying one CA 302 R to see if it has any effect at all. Unfortunately they are about $45.00. Hope this might be of some help.
pgershon Posted October 9, 2020 Author Posted October 9, 2020 I ordered this to try: https://www.mouser.com/datasheet/2/212/F3294_PHE840E-1101083.pdf Also this:
garybixler Posted October 9, 2020 Posted October 9, 2020 42 minutes ago, pgershon said: I ordered this to try: https://www.mouser.com/datasheet/2/212/F3294_PHE840E-1101083.pdf Also this: Just for comparison the CA 302 R is a 2.2 uf capacitor.
Recommended Posts