Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations MintJulep on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
motor overload tripping classes eg class 10,15,20
- Thread starter phasey
- Start date
jbartos
Electrical
- Jan 15, 2000
- 6,440
Suggestion: Visit
etc. for more info
page 13 shows Overload Relay Classes, their trip time versus multiple of full load current
etc. for more info
page 13 shows Overload Relay Classes, their trip time versus multiple of full load current
phasey,
The tripping curves actually have at least 3 detemining points, I believe at 1x FLA, 2x FLA and 6x FLA. It is only the 6x number that is used to classify the curve for simplicity. It is also important to understand that the curve class is based on a MAXIMUM tripping times, but protection devices can pass if they trip sooner. This can of course mean that nuisance tripping may occur, and frequently does with cheap overload devices. Also,
"Venditori de oleum-vipera non vigere excordis populi"
The tripping curves actually have at least 3 detemining points, I believe at 1x FLA, 2x FLA and 6x FLA. It is only the 6x number that is used to classify the curve for simplicity. It is also important to understand that the curve class is based on a MAXIMUM tripping times, but protection devices can pass if they trip sooner. This can of course mean that nuisance tripping may occur, and frequently does with cheap overload devices. Also,
"Venditori de oleum-vipera non vigere excordis populi"
Hello phasey,
There is an FAQ on this subject in the FAQ section if you are interested. There is also a very good discussion thread at
I hope these will answer any questions you may have on the subject.
Regards,
GGOSS
There is an FAQ on this subject in the FAQ section if you are interested. There is also a very good discussion thread at
I hope these will answer any questions you may have on the subject.
Regards,
GGOSS
jbartos
Electrical
- Jan 15, 2000
- 6,440
Comment: The intent of the thermal overload is to have its I-t characteristic close to the motor starting characteristic and as far as possible from the motor thermal durve. Therefore, different classes have been developed to protect the motor sensitively.
jbartos
Electrical
- Jan 15, 2000
- 6,440
Comment. The overload protection is not exactly accurate protection considering the overload sensors (heaters) principles of operation. That is why there are so many malfunctions happening in areas of the motor overload sensing and relay protection.
Amplifying buzzp's point, we have had excellent results using Sprecher & Shuh solid state O/L's. Our applications are chiefly on rotary electric vibrators, which are quite prone to O/L events, due to frequent borderline abuse operation. We also have saved a few compressor motors by using these, also a tough app.
BK
BK
jbartos
Electrical
- Jan 15, 2000
- 6,440
Comment: The situation with the solid state relays accuracy is somewhat sporadic. Visit
for:
no solid state overload relay accuracy is seen upfront
for:
Advantages include: ambient insensitivity, greater accuracy,....
///How greater accuracy is the greater accuracy?\\etc. for more info
for:
no solid state overload relay accuracy is seen upfront
for:
Advantages include: ambient insensitivity, greater accuracy,....
///How greater accuracy is the greater accuracy?\\etc. for more info
Jbartos,
I am not sure what those two links are showing us or telling us. The solid state relays are very accurate devices assuming a sinusoidal waveform. However, there accuracy may worsen if your talking about hacked up waveforms or distorted waveforms. This is because they generally operate assuming a pure sine wave and use the averaging of peak detection to come up with the rms value of current. They generally do not respond to the real rms current such as a temperature operated OL. I am sure there are some that actually have an rms responding circuit but not many. So yes, in the case of distorted waveforms, solid states do suffer some accuracy, in general. But not always.
I am not sure what those two links are showing us or telling us. The solid state relays are very accurate devices assuming a sinusoidal waveform. However, there accuracy may worsen if your talking about hacked up waveforms or distorted waveforms. This is because they generally operate assuming a pure sine wave and use the averaging of peak detection to come up with the rms value of current. They generally do not respond to the real rms current such as a temperature operated OL. I am sure there are some that actually have an rms responding circuit but not many. So yes, in the case of distorted waveforms, solid states do suffer some accuracy, in general. But not always.
jbartos
Electrical
- Jan 15, 2000
- 6,440
Comment on the previous posting: I agree that the posting do not reveal much about ther solid state relay accuracy. That is why they were posted.
I have been looking for specific accuracy values or ranges, e.g. in tech spec:
and there appear to be none.
I have been looking for specific accuracy values or ranges, e.g. in tech spec:
and there appear to be none.
We are right now finalizing the design for production on a new SSOL (which by the way will do exactly what the OP wants) so I can speak directly to the accuracy issue. When duplicating the functionality of a bimetal OLR, we took measurements of many manufacturers' units in operation, finding the inacuracy to be astounding, i.e. as high as +- 20% in some. Fortunately the only thing relatively consistent was the fact that they rarely exceeded the maximum trip time for their class. Most tripped earlier and many would even come under the heading "nuisance" if you asked me. Still, better safe than sorry.
In the SSOL world, accuracy is dead on perfect as far as the digital processes go. When the algorithm says to trip at a particular level in a set time, it does it. The inaccuracy in the system comes from the input measurement method and the sampling rate. We are using 2% class CTs in ours so there is at least a +- 2% risk there. We have played with the sampling rate a bit and found that beyond a certain level, easily atainable by even the lowest cost components, there is no significant loss of accuracy. So even the cheapest of SSOLs should be more accurate that the majority of bimetals. Of course anyone can make crap, but without an economic incentive to do so, why would they make the sizeable investment in bringing a product like that to market?
For buzzp, you are dead on about the distorted waveforms. A version for use behind VFDs will use Hall Effect transducers and I have yet to see accuracy results on those, but I am told they will be similar. The sampling gets real tricky though and we need to use a comparatively high rate to avoid adding error. That version is still in the skunkworks.
"Venditori de oleum-vipera non vigere excordis populi"
In the SSOL world, accuracy is dead on perfect as far as the digital processes go. When the algorithm says to trip at a particular level in a set time, it does it. The inaccuracy in the system comes from the input measurement method and the sampling rate. We are using 2% class CTs in ours so there is at least a +- 2% risk there. We have played with the sampling rate a bit and found that beyond a certain level, easily atainable by even the lowest cost components, there is no significant loss of accuracy. So even the cheapest of SSOLs should be more accurate that the majority of bimetals. Of course anyone can make crap, but without an economic incentive to do so, why would they make the sizeable investment in bringing a product like that to market?
For buzzp, you are dead on about the distorted waveforms. A version for use behind VFDs will use Hall Effect transducers and I have yet to see accuracy results on those, but I am told they will be similar. The sampling gets real tricky though and we need to use a comparatively high rate to avoid adding error. That version is still in the skunkworks.
"Venditori de oleum-vipera non vigere excordis populi"
![[upsidedown]](/data/assets/smilies/upsidedown.gif "[upsidedown] [upsidedown]")
In general, I have found that if no easily obtainable accuracy specs are available then the device is probably not worth buying.
Jraef, is your new SSOL using a RMS method or are you calculating this based on the peak or average value? Just curious. As far as the VFD OL's, I have heard (not confirmed nor tested nor thought about much) that depending on the CT and the carrier frequency of the drive, the waveform that comes out of the CT would already be filtered (smoothed) enough to not require any special sensors (hall effects or other). Since the motor smooths out the same signal, then the argument is no special sensor is required. It looks as though you work for a drive manufacturer. I would be interested to hear if the above statements hold water compared with any testing you have done. What methods do you use, presently, to measure the current out of the drive? Thanks.
Jraef, is your new SSOL using a RMS method or are you calculating this based on the peak or average value? Just curious. As far as the VFD OL's, I have heard (not confirmed nor tested nor thought about much) that depending on the CT and the carrier frequency of the drive, the waveform that comes out of the CT would already be filtered (smoothed) enough to not require any special sensors (hall effects or other). Since the motor smooths out the same signal, then the argument is no special sensor is required. It looks as though you work for a drive manufacturer. I would be interested to hear if the above statements hold water compared with any testing you have done. What methods do you use, presently, to measure the current out of the drive? Thanks.
- Thread starter
- #18
- Status
Similar threads
- Locked
- Question
- Locked
- Question
- Locked
- Question
- Question