24VDC Control Circuit Intermittent Fault

[Timpi Electric Solutions]

I would replace the DC power supply. I had a perplexing case where every few hours the system would burp. Unfortunately, it was a corn chip bagger that would then dump a bunch on the floor. I put my VOM on the power supply, set for minimum, and was able to prove it that way.

I will likely end up doing that, unless I find a better suggestion here. It just drives me up the wall to not KNOW what the cause of the problem is.

 

[Timpi Electric Solutions]

You can make a cheap one out of just about any low end PLC.

Do you just wire series inputs from the field devices to the PLC, then write some sort of program that monitors and records input states?

 

[petersonra]

Do you just wire series inputs from the field devices to the PLC, then write some sort of program that monitors and records input states?

Not in series, parallel.

The program is pretty simple once you decide what you want to do.

 

[EC Dan]

A PLC and program to look at inputs/outputs is simple, but the extra stuff you'd need to actually record data increases the complexity and cost. You're going to need an actual computer connected to this thing somewhere on the network to store the data in a CSV file or database.

 

[petersonra]

A PLC and program to look at inputs/outputs is simple, but the extra stuff you'd need to actually record data increases the complexity and cost. You're going to need an actual computer connected to this thing somewhere on the network to store the data in a CSV file or database.

depends on what you want to record. Most plcs can record a small amount of data and most of the time that is all you need. Plus, if I was doing something like this it would already have a PLC on it so there would be no reason to add something external.

 

[petersonra]

This 3 wire start/stop circuit is for a PF40 but most drives can do the same kind of thing. If you want to add an estop, just wire NC contacts from the estop in series with the stop PB. I rarely if ever would use an external supply for something as simple as what you are doing. It is another complication that serves no useful purpose.

 

[Timpi Electric Solutions]

A PLC and program to look at inputs/outputs is simple, but the extra stuff you'd need to actually record data increases the complexity and cost. You're going to need an actual computer connected to this thing somewhere on the network to store the data in a CSV file or database.

The location is extremely dusty and isn't really somewhere that I'd want to leave my laptop. Are there purpose-built devices that can be connected, initiated, and left for a period of time? The panel isn't networked, either.

 

[Timpi Electric Solutions]

This 3 wire start/stop circuit is for a PF40 but most drives can do the same kind of thing. If you want to add an estop, just wire NC contacts from the estop in series with the stop PB. I rarely if ever would use an external supply for something as simple as what you are doing. It is another complication that serves no useful purpose.

View attachment 2559858

A thermostat and electric solenoid valve are also fed by the power supply. Plus, this customer continually makes changes or adds production equipment, so I have habitually built panels with enclosures that were larger, feeders that were oversized, and included more capability than was strictly necessary.

 

[petersonra]

The location is extremely dusty and isn't really somewhere that I'd want to leave my laptop. Are there purpose-built devices that can be connected, initiated, and left for a period of time? The panel isn't networked, either.

I am not sure why you would want to leave your laptop.

There are plenty of small and cheap dataloggers that could be used, depending on what you need.

Try googling battery powered datalogger.

 

[gar]

220316-1735 EDT

beck693:

As I previously indicated it is likely that your problem is located in a small section of your circuity. As a test tool remove your LED indicator. In its place put the LED diode of a 4N35 in its place. This won't change the series impedance much. 10 mA into the 4N35 will provide a good signal output. For a load resistor on the output side of the 4N35 pick a value that doesn't let the output of the 4N35 output transistor drop lower than about 2 to 3 V with 10 mA on the input. Use a 12 V DC supply for the output side of the 4N35. This will allow you to somewhat measure the relay load current, but it is not linear. Response time relative to your needs will be very good.

You can have an identical circuit to simultaneously monitor the input voltage at any point. For this voltage monitor you pick a series resistor to the LED input to give you about 10 mA to the LED with your expected maximum voltage. You may want to put a 1N4148 reversed biased diode across the 4N35 input diode to prevent accidental high voltage reverse voltage to the 4N35.

.

 

[gar]

220318-1231 EDT

back693:

If you were to use optical couplers to study the circuit. then it might be best to make the coupler that replaces the LED pilot light be in a position very close to the relay coil. This way the coupler used to sense voltage at various points could be anywhere up to close to the relay without the current to the voltage coupler being seen by the current coupler. Alternatively you could reduce the current required for the voltage coupler to a lower value with a larger resistor to reduce its required current to possibly 1 mA.

The use of an optical coupler for the voltage measurement is only necessary if you need to isolated the common lead to the monitoring device from the common of the monitored circuit.

.

 

[gar]

220319-1711 EDT

beck693:

Since REL5 is not dropping out when REL4 drops out this means your problem does not exist in any circuitry common to both 4 and 5. The only area where dropout can come from is where the wire to SP101.41 starts, and where the wire 0009 ends.

I have an additional way you can get measurements from your circuit.

You need one sensor directly across the coil of REL4. We need a sensor that requires little current compared to the coil current of REL4. Apparently that coil resistance is around 12/0.01 = 1200 ohms. I have reed switch 24 V coils with a resistance of 2500 ohms. This not as low as I would like, but might be not too much additional load.

Another identical reed relay circuit can be moved from point to point to find out where dropout does not occur

Consider what one of my 24 V coils with a random reed can do. Experimentally it pulls in at 2.0 mA, and drops out at 1 mA, or pulls in at 5 V and drops out at 2.5 V. The coil resistance is 2500 ohms. With no change to the coil, the relay load resistance could be moved up to about 3750 ohms, or even somewhat higher with added external series resistance.

Reed relays have a response time better than 1 milliseconds.

.

 

[Fishbrain]

Bold Italics are mine.



I'm having a troubleshooting error that I just can't seem to figure out. Several years ago, I designed and built an industrial control panel for power six corner conveyors for a customer. Six Allen-Bradley Powerflex 523 drives power 1HP 480VAC motors. I used a 24VDC 2-wire SRC non-reversing control circuit with an external power supply. A SPDT relay is held closed to provide the RUN command to the VFDs. Each motor is individually controlled with its on set of operators.



Judging from your narrative, I can infer that the schematic diagram is the AS BUILT drawing that the system is based on.. . . although I noticed that some needed information is (are) missing. You have two VFDs presently installed and you added four VFDs.

Initially, only two of the six conveyors were installed and used. The production manager knew that changes would be made, so the conveyor lines were not finalized. The two conveyors worked well with no faults for well over a year. Several months ago, the final four powered conveyors were installed. Previously, the plan had been for a single set of operators and a single potentiometer to control the speed and operation of VFDs 3-6, but the company decided to split them up and have each individually controlled. I added more pushbutton enclosures, added a few relays, and re-wired part of the panel.



The sequence of operation of the six conveyors is what I’m missing. I don’t know if leaving them out in your drawing was done purposely. It maybe because you felt they are not important when asking for guidance?

Another aspect in the operation of those conveyors-- is the SEQUENCE of operation.
Do all conveyors operate sequentially as in Conveyor 7 is fed by Conveyor 6, Conveyor 6 is fed by Conveyor 5 while Conveyor Conveyor 5 is fed by Conveyor 4. . . . so forth and so on?

Do they operate in conjunction with rest of the conveyors?

This question my sound mundane but since you added the four conveyors using the existing control power--this does have some impact on the electronics circuitry of the VFDs.

Bear with me and I’ll explain albeit briefly.
In Electronics there are seven different GROUNDS.

1. Analog ground

2. Digital ground

3. Common ground
4. Chassis ground

5. Power ground

6. Signal ground

7. Earth ground

Electronic Engineers have a contemptuous kinship with Electrical Engineers when designing control circuits. lol
For Electrical Engineers . . . ground is ground—bury your dead down there.

Electronics Engineers "meme" for ground as derived from mantra:

Ground is where POTATOES and CARROTS belong. lol

Since making the changes, I've had a problem with two of the conveyors; VFD 2 (one of the initial two VFDs) and VFD 4 (one of the recent additions). Both of these conveyors will occasionally stop, but will start again as soon as the operator presses the START pushbutton. The relay will de-energize and open the N.O. contact, breaking the circuit and no longer providing the RUN command to the VFD. At first, I was only told that it was a single conveyor (VFD 2) that was stopping. As it had been running fine for quite some time, I assumed that it was simply a relay coil with a unexpectedly short life. I replaced the coil and assumed that the problem would be corrected.

It did not resolve the error, and I was then informed that a second conveyor (VFD 4) was also behaving the identically; it would intermittently stop, but would start without fail as soon as the START operator was pressed.

If you interconnected these “ grounds” especially in the VFD input. . . in the VFD side you could expect funny results.
Going into the “MOLECULAR LEVEL” is not the premise of this forum.

I will be talking to myself if I did.

Some appropriate websites that could explain this . . . .but you are only concerned of the adverse effect of improperly “grounded “ electronics circuitry.
Remember that a complicated problem like you have could be something not mentioned here. Just one call doesn’t explain them all.

The control panel has a compressed air panel cooler, controlled by a 24VDC thermostat and 24VDC solenoid valve. I thought that perhaps when the thermostat was opening the solenoid valve, it was causing a voltage drop and de-energizing some of the more voltage sensitive relays. Extensive manipulation of the valve did not replicate the problem, and the problem remained even when the thermostat was turned off. Additionally, the power supply is 100W and neither read a voltage drop when the solenoid opened, or an excessive surge in load. I have been unable to find the electrical specifications of the solenoid valve, but no other components within the control panel amount to a very significant DC load. At no point have any of the VFDs faulted, they simply stop receiving the RUN command and the conveyor motor is stopped.

My most recent solution was to replace the N.O. and N.C. contact, and the LED indicator in the the pushbutton enclosure for VFD 2, as well as blow out the enclosure with compressed air, but after returning several weeks later, the fault remains today. My mentor is at a loss, as well.

I am at an absolute loss as to the next step for troubleshooting or most likely repair. I've been attempting to attach images of my schematics, but am having trouble properly compressing the images small enough to make the server happy.

Any guidance or insight would be most appreciated.

Good Luck and may you prosper.

Fish

 

[V02maxed]

I spend a lot of time out on the production floor troubleshooting circuits and problems like this. I work in a food plant where there is a sanitation process every week, so much of the equipment and electrical get "watered" occasionally. Most intermittent problems like this come down to a bad/loose connection in a terminal block or what I see many times is a corroded DC contact in a switch/relay. If it were me, I would meter through my estops since you are using them to control the 24vdc and I would make sure you are getting at least 22-23vdc at the first RELY 7 coil. maybe even have someone jiggle or move the Estop button around while you meter. Estop contacts are sensitive, and any dc corrosion could be dropping voltage, you might be right on the edge of dropping the coil out with a loose or bad connection in one of the Estop enclosures. Since you said it's happening to BOTH VFDS, it must be upstream, leading me to the Estops and RELY 7. The start/run inputs on a vfd draw milliamps, there should be no reason the power supply cannot support this circuit.

 

[gar]

220503-1519 EST

beck693:


Reading back a little I see you are working at 24 V and low currents.

You are apparently working at the 24 V level, and not 120 V. What is the contact material on switches and relays? If it is silver-cadmium-oxide you may have a problem at 24 V and lower. For low voltage and / or low current circuits you want to use at least pure silver contacts, and possibly gold over nickel. Silver-cadimum-oxide is a more arc resistant material for switching, but may develop a high surface resistance at lower voltage that is not easily broken down. Usually the same relay with silver-cadmium-oxide contacts will have twice the current rating as a pure silver contact for use at 120 V.

.

 

[Timpi Electric Solutions]

I've had to step away from this project for awhile due to more urgent jobs elsewhere, but I appreciate everyone's input. At some point, if things ever slow down a bit for me, *laughs nervously* I will get back to it.

 

[Jraef]

I’ve chased similar issues in the past, on more than one occasion it turned out to be the power supply. Many “inexpensive” Chinesium power supplies have a built-in crowbar circuit to protect themselves by turning off if you attempt to overload them, even briefly. My guess here is that your solenoid valve is occasionally “sticking”, meaning it is mechanically not allowing the solenoid to fully close as it normally would, which results in high current draw. DC solenoids don’t have an “inrush current” like AC solenoids do, but if the plunger takes too long to close, the current will increase until it does. Your test of rapidly exercising the solenoid valve may not duplicate that scenario because you haven’t duplicated the “stiction” that the system may occasionally see. I personally would just put in a better quality DC power supply (one with a higher “surge current” rating) or a larger one.

 

[petersonra]

I’ve chased similar issues in the past, on more than one occasion it turned out to be the power supply. Many “inexpensive” Chinesium power supplies have a built-in crowbar circuit to protect themselves by turning off if you attempt to overload them, even briefly. My guess here is that your solenoid valve is occasionally “sticking”, meaning it is mechanically not allowing the solenoid to fully close as it normally would, which results in high current draw. DC solenoids don’t have an “inrush current” like AC solenoids do, but if the plunger takes too long to close, the current will increase until it does. Your test of rapidly exercising the solenoid valve may not duplicate that scenario because you haven’t duplicated the “stiction” that the system may occasionally see. I personally would just put in a better quality DC power supply (one with a higher “surge current” rating) or a larger one.

As good a guess as any.

I would not be blaming the power supply for a solenoid that is failing though.

 

[gar]

220714-1205 EDT

DC coil relay coils and solenoids never have a higher DC current greater than defined by the resistance of the coil. There may be an extremely short initial capacitive current, but this is probably under a fraction of a millisecond. Further the transient curve for input current to an LR series circuit is an exponential curve starting at 0 current assuming the initial current before application of power was 0. Basic electrical theory.

For a continuously excited DC coil the device must be designed for this power dissipation. But it also means you won't burnout the coil if the armature does not fully transfer.

In contrast most AC power relay coils will burnout if the armature does not fully close in an short time.

.

 

[SceneryDriver]

I’ve chased similar issues in the past, on more than one occasion it turned out to be the power supply. Many “inexpensive” Chinesium power supplies have a built-in crowbar circuit to protect themselves by turning off if you attempt to overload them, even briefly. My guess here is that your solenoid valve is occasionally “sticking”, meaning it is mechanically not allowing the solenoid to fully close as it normally would, which results in high current draw. DC solenoids don’t have an “inrush current” like AC solenoids do, but if the plunger takes too long to close, the current will increase until it does. Your test of rapidly exercising the solenoid valve may not duplicate that scenario because you haven’t duplicated the “stiction” that the system may occasionally see. I personally would just put in a better quality DC power supply (one with a higher “surge current” rating) or a larger one.

I've also seen cheap power supplies hiccup and crowbar when driving DC solenoids when there is no freewheel diode installed across the solenoid coil. Most switchmode power supplies REALLY don't like reverse power.


SceneryDriver