3 Wire Paper machine, 500mpm(max) uses analogue sectional DC drives & PLC for draw control. It is required to do away with the PLC and speed regulator due to lack of spares. A new PLC will handle draw, HMI, AND speed regulation. Tacho is 300 ppr for 1500rpm motor. Is speed regulation of 0.3mpm achievable? comments,concerns
Eng-Tips is the largest engineering community on the Internet
Intelligent Work Forums for Engineering Professionals
-
Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
Drive revamp 1
- Thread starter astroid
- Start date
Assuming the the machine travels at 500 mpm at 1500 rpm then you have 7500 pulses per second or 7.5 pulses per millisecond. You will need a high speed pulse input module to bring your tach signal in and the scan time on your PLC will need to be small or will need to have a function where the speed control signal can operate with priority (maybe multiple times) over other processes. Some PLC's can have multiple processors so you could dedicate a processor to just speed control.
A lot of paper machines in the US are PLC controlled and I don't see why you shouldn't be able to do so. Just pick the right PLC and I/O. I assume that you have been able to control to 0.3 mpm with the existing setup so there is no problem with the DC drives responding.
This question is probably better suited to the PLC forum.
A lot of paper machines in the US are PLC controlled and I don't see why you shouldn't be able to do so. Just pick the right PLC and I/O. I assume that you have been able to control to 0.3 mpm with the existing setup so there is no problem with the DC drives responding.
This question is probably better suited to the PLC forum.
-
1
- #4
Skogsgurra
Electrical
Hehe Scotty..
I woke up a lot earlier than this. But phone call sent me digging out an old pile of papers. So haven't had the time until now.
I agree with gepman. It is doable, with the right equipment. A well designed drive can make an old machine shine. But it can not do miracles. Any torque or draw variations should be recorded and documented before the rebuild. The reason for them should also be found (and hopefully remedied) so that your new drives don't get the blame if there is continued poor performance. That is a situation that isn't too uncommon.
Paper machine drives in the eighties used to have rather poor resolution encoders. 256 or 500/512 PPR were standard values at that time. And that made a typical 0.005 percent speed control possible. Static, long-term accuracy, that is. Dynamic speed deviations depend on other factors like inertia and dynamic load variation. The trick to get fast and accurate speed measurement is to do reciprocal speed measurement over a measurement interval range and then use an running average, usually a first order IIR filter to get a noise-free speed signal. Many middle-sized and all high-end PLCs have that available. It can even be found inside many modern digital thyristor rectifiers.
The .3 meters/minute regulation at 500 m/min is around 0.06 percent and not a very difficult spec, if it is a static one. For short term deviations, it is common to see less than 0.2 percent-seconds. The percent-second unit is the area in percent*seconds that the speed deviates from set speed when a step torque variation being 50 percent of the rated torque is applied to the drive. since most drives has a 50 % overload capacity, it doesn't matter if the torque step is applied in negative or positive direction. Even if the drive is working near full load.
TAPPI has some good papers on this. And so has IEEE.
Regarding encoder resolution. There is a rule-of-thumb (Siemens) saying that PPR shall be more than 275/minimum RPM for good runability at low speeds. So, in your case, if the machine runs between, say, 200 and 500 m/min, the PPR needed to run at all at 200 m/min would be something like 1 PPR. This, of course, means that quadrature decoding is used so what you really get is four bits of speed information per revolution. In the 1500 RPM case, that is 600 RPM at 200 m/min - or 10 RPS. So you will have an updated speed information 40 times per secoond and that is usually all you need to run machines with high inertia.
Wire sections do not have a high inertia. They need better speed measurement. But, your 300 PPR is 300 times better than the 1 PPR necessary for a high-inertia section. So it is OK also in this respect.
I have done such rebuilds on several machines. Including winders. No probs.
Gunnar Englund
--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...
I woke up a lot earlier than this. But phone call sent me digging out an old pile of papers. So haven't had the time until now.
I agree with gepman. It is doable, with the right equipment. A well designed drive can make an old machine shine. But it can not do miracles. Any torque or draw variations should be recorded and documented before the rebuild. The reason for them should also be found (and hopefully remedied) so that your new drives don't get the blame if there is continued poor performance. That is a situation that isn't too uncommon.
Paper machine drives in the eighties used to have rather poor resolution encoders. 256 or 500/512 PPR were standard values at that time. And that made a typical 0.005 percent speed control possible. Static, long-term accuracy, that is. Dynamic speed deviations depend on other factors like inertia and dynamic load variation. The trick to get fast and accurate speed measurement is to do reciprocal speed measurement over a measurement interval range and then use an running average, usually a first order IIR filter to get a noise-free speed signal. Many middle-sized and all high-end PLCs have that available. It can even be found inside many modern digital thyristor rectifiers.
The .3 meters/minute regulation at 500 m/min is around 0.06 percent and not a very difficult spec, if it is a static one. For short term deviations, it is common to see less than 0.2 percent-seconds. The percent-second unit is the area in percent*seconds that the speed deviates from set speed when a step torque variation being 50 percent of the rated torque is applied to the drive. since most drives has a 50 % overload capacity, it doesn't matter if the torque step is applied in negative or positive direction. Even if the drive is working near full load.
TAPPI has some good papers on this. And so has IEEE.
Regarding encoder resolution. There is a rule-of-thumb (Siemens) saying that PPR shall be more than 275/minimum RPM for good runability at low speeds. So, in your case, if the machine runs between, say, 200 and 500 m/min, the PPR needed to run at all at 200 m/min would be something like 1 PPR. This, of course, means that quadrature decoding is used so what you really get is four bits of speed information per revolution. In the 1500 RPM case, that is 600 RPM at 200 m/min - or 10 RPS. So you will have an updated speed information 40 times per secoond and that is usually all you need to run machines with high inertia.
Wire sections do not have a high inertia. They need better speed measurement. But, your 300 PPR is 300 times better than the 1 PPR necessary for a high-inertia section. So it is OK also in this respect.
I have done such rebuilds on several machines. Including winders. No probs.
Gunnar Englund
--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...
- Status
Similar threads
- Locked
- Question
- Locked
- Question