Problems with Phase to Phase Heating Elements

I have a machine that only has 3-phase 220Vac (and plant's 3-phase 480Vac). There are 2 heating elements. Element1 is powered by Phase1 and Phase2 (220Vac phase to phase). Element2 is powered by Phase2 and Phase3. Each element draws 22 amps when powered (using PID temperature control). The fact that elements are floating causes a lot of maintenance problems, so I'm wondering if it would make sense to rewire it so that Element1 is powered by Phase1 to Ground (127Vac, 12.7A), Element2 powered by Phase2 to Ground. (I believe target temperature is low enough that it can still be achieved despite power downgrade.) Note that the machine also has four 220Vac 3-phase ac motors, among other things. Question is: will this design change create massive problems? (Probably yes.) What problems will I see? Should I just buy a 380 25A transformer, so I can get two 220 phase to ground?
 
>... The fact that elements are floating causes a lot of maintenance problems, ...

Please validate "problems".

Additionally, having 127V line to "neutral" suggests your system is a wye configuration. You didn't mention where you are located, but here in the U.S., a system in that voltage range and configuration is required to have a grounded neutral, with only a few exceptions. If your neutral is grounded as I suspect, the elements are not floating and not causing problems as purported.

I'm with Phil... you need to provide at the very least a one-line diagram... and it appears you need to include the power source and any grounding thereof.
 
Runescape... Thanks for the sketch!


I still don't understand how an ungrounded/unearthed supply influences Heater maintenance. Are you experiencing intermittent failures of SSR control elements? Or is the supply experiencing phase-to-ground voltage excursions?

If the latter, then you can certainly install a derived-ground using several techniques, which will eliminate the intermittent faults. But, not until you provide additional information about how the SSR is triggered.

Caveat: Do not connect the Heater elements to PE under any circumstance.

Phil
 
S
Running heaters with hots only and no grounded conductor shouldn't be causing you any problems; it's often done that way. Think about it. any high KW heating system (say 15-20KW and up) will naturally be 3-phase, and none of the three phases is normally grounded at the transformer. Typically, the center point will be grounded, creating a neutral (grounded) conductor, which although not connected to most loads, will give the ungrounded conductors a stable voltage to ground (277 in a 480 system, 120 in a 208 system). Normally, the metal case of the heater is grounded for safety, but it's not part of the normal design current path. Even with a quartz or teflon sheathed heater, there is some sort of grounding system for safety, though obviously not connected directly to the sheath since it's non-conductive. You're getting bad advice or you have a very atypical situation in your plant.
 
I'll try to explain the problem.

The machines in question perform a fast chopping motion and need to be cleaned and parts replaced, etc. several times a day. There's motion and human workmanship involved, so a lot of things can go wrong.

Most common problems I see are

1. Solid State Relay gets stuck on. This happens way more on the 2-phase setup, than the line-to-neutral setup. I'm not sure if it's because both sides of the SSR are hot, or because of something else .

2. Sometimes a section of the heating element will go bad, or maybe a point on the wire insulation will rub away exposing copper to the machine's frame (earth/neutral). This basically means current has a path phase-to-ground now, and the side with the SSR will get about half power when SSR closed; the other side will get half power all the time. This causes a lot of problems! (If instead the setup was phase-to-neutral, the section of element between neutral and the ground will just not be powered. This would still a problem, but not as bad.)

For both the above cases, I have some countermeasures in place, such as a CT switch per circuit (so 2 of them) to catch if the SSR is stuck closed. Also a high temperature alarm that kicks out the contactor (denoted as "on/off" in my sketch).

I suppose a full countermeasure package would be 2 CT sensors per circuit (so 4 total), 2 SSRs per circuit (4 total), and maybe 2 more temperature probes as backup readings. Would have to add analog input card, another TC card, etc. And all of this just because the machine uses the 2-phase design? Why not just redesign it to be phase-to-neutral like the other machines (which have a 3-phase 380 transformer, so can send 220 phase to neutral)?

That's basically my line of thought and why I ask this. Can I rewire my existing setup phase to ground? Will my ac motors start acting weird if now only 2 phases are drawing power for heat? Will my heating elements have problems when going from 220V to 127V (and longer pulses of power)?

Or... do I have to just buy several 3-phase 380Vac transformers (one for each misbehaving machine) so they'll behave like the good machines?
 
> I still don't understand how an ungrounded/unearthed supply influences Heater maintenance. Are you experiencing
> intermittent failures of SSR control elements? Or is the supply experiencing phase-to-ground voltage excursions?

Both. Explained in my other reply. (http://www.control.com/thread/1422054751#1422257455)

>If the latter, then you can certainly install a
>derived-ground using several techniques, which will
>eliminate the intermittent faults. But, not until you
>provide additional information about how the SSR is
>triggered.

The elements are to heat an object and keep it at a constant temperature. An RTD on the object. PLC has a temperature controller card. PID loop. Card's binary output goes to the SSR which is NO. Cycle Time I think is 5 seconds, more or less. Typically there will be an output every 5 seconds, but pulse duration will not be very long. Maybe half a second. The rest pretty much matches my sketch.

> Caveat: Do not connect the Heater elements to PE under any circumstance.

Doh. So basically the redesign will not work? Please explain. Please also explain the derived ground.
 
B
I don't see how you gain anything from most of what you are proposing. Unless the equipment is just not appropriate for what is being done, it should work fine.

SSR rarely fail. If they do, it is best to figure out why they are failing rather than applying bandaids. Perhaps they are misapplied.

I don't think you can change the heater voltage from 220V to 127V. That would result in a loss of about 2/3 of the heat output.
 
> For the present configuration what are the phase-ground voltage measurements when a "Heater-element" is On? And when it's Off?

This data is from a machine that was not running. Running machines have what look to be same average values, but swinging all over the place. Note that 3 of the ac motors are for dancers, so speed is never constant (VFD driven by analog prox sensor). One is for a carriage motor that is sometimes still, sometimes moving (VFD with 4 speed presets). The rest are just conveyers with direct drive.

Element1 SSR side
105 when off, element2 off
115 when on, element2 off
102 when off, element2 on

Element1 non SSR side
105
107
102

Element2 SSR side
138
102
138

Element2 non SSR side
138
139
138
 
B
I am not sure what you mean by the numbers you posted. Are these voltages? What are you measuring them with respect to. What are you measuring them with (a digital meter that is high impedance will give some funky numbers with SSR).

Unless the SSR are not rated for the voltages and currents they are running at, there is no reason for them not to work. Are they mounted on a proper heat sink?

It is very hard to diagnose this kind of problem when a lot of irrelevant information is posted but not much in the way of useful data.

It is highly unlikely that the motors or the VFDs are affecting the SSR.

Is there some way you can sketch up the circuits involved and post it somewhere that we can see it?
 
S
If the heaters are subject to mechanical damage, why not fix the mechanical design instead of trying to make the electrical design tolerate it?

The fact that putting 380V transformers in front of the machines is on the list of potential fixes makes it sound as if you're trying to run these machines on other than the original design voltage, which is not unheard of, but it sounds as if you don't have the skill set in house to accomplish that conversion. Is that what you're trying to do? If so, I'd suspect that as the cause of problems long before running heaters phase-to-phase, which is done all the time.

If humans are potentially exposed to leakage current from the heaters, and/or it's a wet environment, you should probably have miliamp level GFCI protection on the heater circuits.
 
> I am not sure what you mean by the numbers you posted. Are
> these voltages? What are you measuring them with respect to.
> What are you measuring them with (a digital meter that is
> high impedance will give some funky numbers with SSR).

I drew a sketch in previous post. The numbers were in reply to PhilCorso's question. The units are VAC, measured from a multimeter, of one side of the heating elements relative to ground. (I have 2 elements, so that's 4 sides). There are 3 measurements per side: first one is when SSR is off (no power through the element), second one is when SSR is on (power going through the element), third one is an extra I threw in (power going through the other element).

> Unless the SSR are not rated for the voltages and currents
> they are running at, there is no reason for them not to
> work. Are they mounted on a proper heat sink?

Can't remember the exact model number, but the SSR is rated for 40A single phase by Omron. It comes with a pretty beefy heat sink.

> Is there some way you can sketch up the circuits involved
> and post it somewhere that we can see it?

https://www.dropbox.com/s/jafg1wxik6eftfw/Screenshot_2015-01-25-15-08-53.png?dl=0
 
>> Caveat: Do not connect the Heater elements to PE under any circumstance.

> Doh. So basically the redesign will not work? Please explain. Please also explain the derived ground.

PE is never a circuit conductor by design. It is only supposed to conduct current during a ground fault, and only long enough for the overcurrent protection device to clear the fault.

N (neutral) is typically a grounded circuit conductor.

"Derived ground" (I believe Phil meant "derived grounded neutral") is typically achieved, for your case, with a delta primary, wye secondary transformer, where the secondary common terminal (X0) is grounded.
 
Runescape... your Phase-to-'Earth' voltage-measurements are indicative of a "floating" system. Probably due to line-earth capacitance. The magnitudes reveal some unbalance.


To eliminate the existence of a "sneak-path" between the SSR modules and earth, I suggest you take an Ohm measurement between the SSR module and earth (with system totally de-energized) and the SSR and Heater elements isolated.

There is merit in substituting an earthed system. Especially if Down-Time cost is high. Additional information is required.

1) How many Machines are involved?

2) Are current machines powered from one source? Or, several 3-phase sources?

3) Have you made comparative measurements of the so-called 'good' machine?

Phil
 
Top