Maxi tronic speed wont read out correctly due to noise on ground..

[ctaylo360860]

I have trouble shot all wiring with engineer from Maxitronic.... The trouble I'm now having is that there is so much noise on the ground that I can not get an accurate read out for the speed moniter. I am not sure the cause and would love for someone to possible point me in an educated direction!

The maxitronic is used to moniter our top drag. it has 4- thermocouples for bearing temperatures, and 4 belt trackers for alignment. it also has a proximity scanner on the end of the shaft that should give an accurate speed reading.. if the drag is running at 100% then its working like it should and we do not have a problem. if the drag reaches 90% of its speed then the there is a 10% alarm that should ring and let someone know the drag is slowing. it will also do this to 80% and set of the 20% alarm.

with the drag running and having no product i have calibrated it, and had no luck of getting a consistent readout for my speed.
with instruction from the engineer i have:
I have removed the drain wire from all other grounds.
I have landed the drain wire on the sh terminal
i have landed the drain wire on the g terminal
i have removed the egc from the terminal block
I have removed drain wire entirely
I have done these all in different combinations and not had much luck, the one thing i did find is this.

I am being told that the target light wants to have a good flashing heartbeat. When i have any part of the drain wire to ground I can not get the target light to do this. instead it gives a faint dimming pulse. also told if there was something miss wired i would not have flashing target light. which is how it was for the last 6 years plus...

This is an existing system, that i did not wire and im beginning to think that there is some issues with grounding and bonding somewhere. I will draw an illustration to show how they had grounding and bonding done at the service, the mcc, transformer, and panels. I thought maybe the VFD's inside the motor control center had something to do with it, but even with them disconnected I'm still getting the sporadic read out... it looks to me as they have all grounded and grounding conductors bonded past the service disconnect.

Thanks again in advance always appreciate the help from this forum!

 

[ctaylo360860]

How equipment in panel room is grounded and bonded...
forgot to draw it, but the maxitronic is fed from the 100A panel with a 20a 120volts circuit.

 

[ctaylo360860]

800a Service disconnect
grounding electrode conductor running to ufer.
1200a breaker feeds mcc1
800a breaker feeds dryer panel

not sure if it’s mis labeled or they wired it like this, but I would of thought the service disconnect would be 1200a and the feeders would be the 2-800a breakers.

 

[zbang]

At the first, it sounds like there's a short somewhere from the drain to one of the signal wires, check for that.

 

[ctaylo360860]

1st pic- mcc1

2nd pic-3 parallel runs of 4 wire feeders from service to mcc main
-Signal wires cross feeder wires in mcc

3rd pic-Feeders to line side of main breaker

4th pic- neutral disconnecting link
has bonding jumper from neutral disconnecting link to insulated Neutral bar and a bonding jumper from Ground bar to Neutral disconnecting link.

5th pic-ground bar

6th & 7th pic- shows isolated ground wire from ground bar to insulated bar

8th & 9th pic0 shows jumper from right side of neutral disconnection link up to insulated Neutral bar

 

[ctaylo360860]

At the first, it sounds like there's a short somewhere from the drain to one of the signal wires, check for that.

I know I have it cut off flush as possible where proximity scanner whip junctions to signal wire...
I will have to make sure it’s isolated at top and bottom and ohm wires and see if I get a reading...
I did get ol on all other wires showing they are good.

 

[ctaylo360860]

1pic- transformer grounded and grounding conductors are bonded
- missing G.E.C.
2pic-100A panel fed from transformer has grounded and grounding conductors bonded
this panel feeds the maxitronic with a120v 20a circuit.

 

[ctaylo360860]

1pic-maxitronic maxiscan
2pic-drain wire i taped to make sure its not touching anything its not supposed to
3pic- maxitronic maxiscan
4pic- diagram showing sensor wiring. we have ez mount (pnp)
5pic- maxitronic maxiscan wiring diagram

 

[petersonra]

My guess is grounding/bonding/shielding is not where your problem lies. It almost never is with these kinds of problems.

How did you determine there was "noise on the ground", and that it is the source of your problems?

 

[ctaylo360860]

My guess is grounding/bonding/shielding is not where your problem lies. It almost never is with these kinds of problems.

How did you determine there was "noise on the ground", and that it is the source of your problems?

Engineer from maxitronic said since speed readout is jumping but everything else seems normal and that heartbeat changes when drain wire is attached he thought it was noise n the grounD.
I found out I could get a scope and possibly put it on the sensor. If I get a clean wave then I’m good f I get spikes I got my problem found.

 

[jim dungar]

Engineer from maxitronic said since speed readout is jumping but everything else seems normal and that heartbeat changes when drain wire is attached he thought it was noise n the grounD.

Classic manufacturer's response. By saying this he has shifted all blame to you the user.
If the shields are causing you grief why are you using them?

The wiring diagram has note saying to maintain separation but the thermocouples are probably mounted into grounded blocks holding the bearings so separation from a noisy ground is not really possible.

 

[ctaylo360860]

Well I will say the is, whoever did the install did a terrible job. They had signal wires mixed up, had a broken wire in j box where proximity scanner wires and signal wire junction.

The thermocouples never gave a temp reading because they weren’t programmed as thermocouple. Same for the belt trackers…
the maxitronic never gave a speed reading or target flash in six years of install…. I got all that to work, but can’t get the damn speed to read accurately
I have spent probably two hours with the manufacturer engineer trouble shooting. He seems like he wants to help me but the conclusion we came up with is the lThe thing that gets me is the thermocouples seem to work it the speed isn’t. Thermocouple and signal wires all in the same raceway…. There is a section were the dial wire runs across the main feeders for thr mcc. Also the is a run of raceway that has the sig/thermo/belt tracker wires that runs parallel to motor runs with only 3 inches of separation, but only for about 20 ft. The strange thing is I who’d think if this was the cause the thermostat couple reading would be funny as well…. Someone in another group suggested a pull down resistor. The engineer of maxitronic mentioned a 1uf capacitor. The guy who suggested the step down resistor said it the Capacitor may cause ossilation and that the resitor would bleed all high/low impedance signals…. Others have suggested an isolatedground
I can say this above my head, but I’m determined to find solution!

 

[petersonra]

You may end up having to pay somebody to come out and spend a few weeks figuring out what's going on. Or the guy might spend a few hours and figure it out. Most places are unwilling to write a 20 or $30,000 purchase order to have a guy come out and sit on a chair in front of the control panel and think about it. But that's probably where you're at if you want to fix the problem.

Or you could randomly change things until it works. That's actually something that works occasionally.

I worked on a punch press one time that had a similar weird glitch. The guy told me that he had been told it was a bad ground and to drive a rod next to the machine and connect the machine ground to the ground rod. I didn't figure it would hurt anything so I didn't argue with him. Didn't really help any either. I told him about all I could do is sit and watch it until it failed again and he said how much would that cost. I think at the time we were charging about $90 an hour. He wasn't willing to pay to have me sit there and watch it until it failed again so I went back to the office. Don't know if they ever resolved it. Maybe they put in another ground rod.

I think I also suggested he rent an oscilloscope so we could see if the encoder pulse is had some kind of weird signal. Didn't want to spend the money on that either. How much you can do when people won't spend money on a problem that can only be solved by spending money.

 

[zbang]

One thing you could do is bring the sensor down to the electronics and connect it with 5' of wire; that'll tell you if the sensor is working. Or, depending on length, temporarily run a new cable up to the sensor. (How much distance is this?)

Putting a 'scope on the sensor leads is an excellent idea, but I'd try moving or replacing the sensor first.

 

[synchro]

... the maxitronic never gave a speed reading or target flash in six years of install…. I got all that to work, but can’t get the damn speed to read accurately
... The thing that gets me is the thermocouples seem to work it the speed isn’t. Thermocouple and signal wires all in the same raceway…. There is a section were the dial wire runs across the main feeders for thr mcc. Also the is a run of raceway that has the sig/thermo/belt tracker wires that runs parallel to motor runs with only 3 inches of separation, but only for about 20 ft. The strange thing is I who’d think if this was the cause the thermostat couple reading would be funny as well….

The thermocouple output is basically a DC signal, and so noise on it could easily be filtered out without affecting the desired signal. But the speed sensor output is a waveform that's changing vs. time, and so there will be a limit on how much noise filtering can be done without affecting the signal as well.

I'm not familiar with the Maxitronic, but I browsed an online document for the MT400 motion controller.
It says:
"Do not run the probe cables in with high voltage [120/440 VAC] wires. Be sure the probe cables are not wrapped around high voltage lines which turn the cable into a coil and induce noise and voltages into the probe."
And so routing of the sensor lines next to power conductors might be contributing to your problem.

The document also says:
"If Target light flashes but is erratic: Check for metal being too close to the probe. Make sure the probe is close enough (3/8”-5/8”) to the target disc"

"The Motion Sensor probe is a metal detection device, it is not a proximity type sensor. This allows probe to pickup targets of both ferrous and non-ferrous metals. The motion sensor sends a pulse signal [12 VAC max] to the MT400 control as it detects MOVING metal within it’s range. The closer metal is to the probe the higher the signal level. The sensing distance usually works best between 3/8” to 1/2”. The Probe can sense metal from all sides of the tip of the probe. The installer must keep" (non-moving) "metal at least 1-1/2” away from the sides of the front half of the probe to prevent it from shielding proper operation."

Perhaps the sensor could be moved closer to the target disk to increase the signal output enough to overcome any noise that may be present. Also, you could check the target disc to make sure that it's not damaged. For example, that it's not missing any metal "nubs" on the periphery of the disk that the sensor needs to detect.

 

[ctaylo360860]

One thing you could do is bring the sensor down to the electronics and connect it with 5' of wire; that'll tell you if the sensor is working. Or, depending on length, temporarily run a new cable up to the sensor. (How much distance is this?)

Putting a 'scope on the sensor leads is an excellent idea, but I'd try moving or replacing the sensor first.

I think this is the method I'm going to try.... I checked it today while trouble shooting some other problems, and when i have the drain wire from the signal and the thermocouple cables isolated i get 40vac to ground on them..... I'm going to go back up and make sure I don't see anything that looks strange with the wiring up top..... im going to bring the sensor and the 5 ft whip down and hook it up right by the control board. i planned on holding it over a cordless drill so that it can track the speed. if i don't get issues there, i guess i know its something with the signal wires going up the leg to the sensor...

 

[ctaylo360860]

The thermocouple output is basically a DC signal, and so noise on it could easily be filtered out without affecting the desired signal. But the speed sensor output is a waveform that's changing vs. time, and so there will be a limit on how much noise filtering can be done without affecting the signal as well.

I'm not familiar with the Maxitronic, but I browsed an online document for the MT400 motion controller.
It says:
"Do not run the probe cables in with high voltage [120/440 VAC] wires. Be sure the probe cables are not wrapped around high voltage lines which turn the cable into a coil and induce noise and voltages into the probe."
And so routing of the sensor lines next to power conductors might be contributing to your problem.

The document also says:
"If Target light flashes but is erratic: Check for metal being too close to the probe. Make sure the probe is close enough (3/8”-5/8”) to the target disc"

"The Motion Sensor probe is a metal detection device, it is not a proximity type sensor. This allows probe to pickup targets of both ferrous and non-ferrous metals. The motion sensor sends a pulse signal [12 VAC max] to the MT400 control as it detects MOVING metal within it’s range. The closer metal is to the probe the higher the signal level. The sensing distance usually works best between 3/8” to 1/2”. The Probe can sense metal from all sides of the tip of the probe. The installer must keep" (non-moving) "metal at least 1-1/2” away from the sides of the front half of the probe to prevent it from shielding proper operation."

Perhaps the sensor could be moved closer to the target disk to increase the signal output enough to overcome any noise that may be present. Also, you could check the target disc to make sure that it's not damaged. For example, that it's not missing any metal "nubs" on the periphery of the disk that the sensor needs to detect.

 

[zbang]

when i have the drain wire from the signal and the thermocouple cables isolated i get 40vac to ground on them

Unless there's an intentional bias on them (highly doubtful) I'd chase that first.

 

[synchro]

im going to bring the sensor and the 5 ft whip down and hook it up right by the control board. i planned on holding it over a cordless drill so that it can track the speed. if i don't get issues there, i guess i know its something with the signal wires going up the leg to the sensor...

If you have trouble picking up the speed of the cordless drill, you could chuck a flat spade bit in it so that the sensor would detect the changing proximity of the metal from the side of the bit.

 

[ctaylo360860]

so the way the engineer had explained it is he didn't think that the thermocouple wires were having an issue because the readouts on the display seem to be pretty constant and not jumping... someone on another forum mentioned that a capacitor could cause an oscillation, that i might try a pull-down resistor since it will bleed off all low level/ high impedance signals. he said just make sure the resistor value doesn't cause excessive current draw on the signal source.... The only place the thermocouple wires and the signal wires actually are with high voltage is in the mcc where the feeders are coming through... i think the minimum required seperation for 300v to 750 is 12 inches. see picture #2. I will try to get a shot of the raceway that has only a 3inch seperation and is a span paralleled about 20ft. I will also get a picture of the cups on the shaft and see if anyone sees anything there that could be causeing a problem. That part all looked pretty good, but im not that familiar with the product. Thank you on educating me on the right terminology I was just repeating what the engineer had called it... thanks for the great reply!