G
Gentlemen,
I would like to use the email posted by Jon Oak (shown below) as a basis for my comments to address the "open" topic that has been so heavily discussed (John, I'll give you my 2 cents on your
application in a separate email). I have been monitoring the conversations concerning the "need" for open systems very closely & it seems to have developed into 2 distinct topics. The first is the need for open architecture and the second is the need to be able to purchase equipment directly over the internet (or from the
manufacturer) at the best price possible.
Over the past several months, many of you have used the term "vendor" with much resentment. I can only assume from the comments & wise-cracks, that many of you view vendors as worthless entities that do nothing but mark up a piece of equipment & sell it for a profit. I work for a "vendor" that considers itself a "value added distributor". I'm sure many of you will get a good chuckle out of that term & will immediately think that the only value I add is someone to beat up when your equipment is late.
I would hope that my previous posts would prove that I have some degree of competence in the motion control industry. I have only been in this industry for 4 1/2 years. However, I have done nothing but motion control for that time. I have seen hundreds of applications & feel I have a tremendous amount of experience to share in terms of what works & what doesn't. Make no mistake about it, I do not claim to be a expert & eat my share of humble pie regularly.
Jon's email proves that providing "open architecture" does not make the system any simpler to use. He's going to use a PC, some operating system (Iet the debates begin over which one he should use), a motion control card, motors, amplifiers, and some sort of mechanics. Further he could use an SST card in the PC to
plug in anybody's I/O. How much more open can you get? Okay, now we've got a truly open system and everybody's happy right? Well who's going to tell him how to make it work? Let's assume he
uses the following:
Galil's DMC-1730 motion control card w/ an Active X toolkit to interface to Visual Basic
Motors from Custom Servo Motors
Amplifiers from Pac Sci
Mechanics from THK
Gearheads from Bayside
How many of you could make this system work? This is not a straw poll, so please don't respond with "I could". The last thing I want is 300 emails telling me how to make it work. I already know how to make it work. Now for those of you that feel comfortable with the first question, how many of those people could get it up and going in less than a week? Once again no response is necessary.
My point to all of this is, that the more open you make the automation world, the more finger pointing you are going to have if you choose to deal directly with the manufacturers. Let's face it, Allen Bradley, Parker, Siemens, etc. wants you to buy their entire system. If you choose to build a system from pieces there is
going to have to be someone to step up & assist you. Integrators do not have the time to become experts on everybody's equipment. But Guy, we don't have to do that.....we'll find one system that we like & standardize on it. If you're an OEM you can do that. If you're an integrator, there's a little thing called "customer spec'd
equipment" that makes that thought process an impossibility.
It seems to me that the desire for open systems strengthens the position that "vendors" (distributors) have. Let's say I sell a
$20,000 system to do what Jon is looking for. Assume that I have a 30% GPM built in to that price. That means that if he could buy direct, he would be able to save $6,000 on capital costs. That's a lot of money right? Well let's look at what that $6,000 gets him:
Pre-sale
1) Engineering time to properly size the motors, drives, & gearheads (torque, speed, regen, inertia matching, etc.)
There are sizing software programs available to automate this task, but some are buggy & if you don't understand what is taken into account when sizing is performed you are at the mercy of the software to select the system for you. The software doesn't tell you everything that it assumes. When you give it a distance to move & a time to move it in, it probably assumes a trapezoidal move profile with a 1/3, 1/3, 1/3 time allocation for acceleration,
constant velocity, and deceleration. What it doesn't tell you is how to program the controller to achieve that move profile. If you program the wrong move profile the system may not work. Also,
what if your motor doesn't have enough speed for the move? Well you could increase the acceleration & deceleration rates to lower the top in speed requirements instead of buying a bigger motor. The opposite could be done to lower the torque & increase the top in speed. Further, many programs don't take regen into consideration or explain the importance of 2) Engineering time to make sure the system will work.
I can't tell you the number of times that I've seen gearheads break because the wrong one was chosen. People constantly try to save money by putting inferior gearheads on servos. A cheap gearhead can be used on steppers because they are either moving or not moving. Servos are always moving (even when they appear to be at rest, they are still servoing / moving). A
servo can destroy a gearhead designed for stepper systems by shaking it apart. If you're in a hot environment or moving constantly at a high rate of speed many gearheads will heat up & transfer the heat back to the motor which could damage or destroy the motor. Many gearheads have the ring gear pressed in. Further the ring gear could be made out of a different material than the housing. If that's the case, the two metals will have a different coefficient of expansion due to heat. This could lead to a lot of problems.
Orientation of the gearhead is also important. If the shaft is facing upward you will need to let the gearhead Will the motor even
mount to the mechanical system? Just because it will give enough torque and speed it doesn't mean it will physically mount. Do I need to clamp the torque of the servo to make sure I don't break
the mechanics? If so, does the control scheme (controller &/or amplifier) have that ability? Can the mechanics handle the forces & moments to give me the life I need? Do I have enough resolution in the system to ensure I can obtain the repeatability I need? Just because the mechanical system has a repeatability spec that
meets my requirements, it doesn't mean that I have enough commanded resolution to obtain that repeatability.
What about compatibility between the drives, motors, and controller? Does the inductance of the motor allow it to work w/ the amplifier? Does the switching frequency of the amplifier pose a problem to the motor? Does the controller have an encoder input great enough to handle the input frequency of the encoder? Is the motion controller's I/O compatible w/ the amplifier's I/O? In other words what if the fault output & enable input on both devices are PNP? That's gonna require extra relays to make the system work.
What about the deratings? Motor's have a torque speed curve that is based on the motor being attached to an aluminum heat sink w/ 10" x 10" x 1/4" dimensions. If you don't have that, you can burn the motor up at rated torque (I've seen it happen). Gearheads are derated based on the duty cycle and the gear ratio.
What about noise? I could write another book on that topic alone.
Post-sale
4) On-site support for hardware setup
We will usually spend at least 1 full day; many times 2 or 3 days to help a customer get up & going.
5) Phone support for programming issues
We have equipment in our office that allows us to simulate programs in the field. This allows us to debug or offer suggestions
to improve the existing program.
6) Wiring schematics to configure the entire system....enough said
7) Reduce finger pointing.
We are the ones that have to deal w/ the manufacturers pointing fingers at each other 95% of the time. I once spent 80 hours in 3 days trying to get a customer's motion controller work w/ the motors & drives I sold him. By the time it was over I knew more about the IMC-123 module than most of the AB guy's did (no, I don't sell AB products).
8) Spares on our shelf
Many manufacturers are cutting costs by becoming JIT manufacturers. This means that normally it will take anywhere from 2 days to 4 weeks to ship.
Is that worth $6,000? I wish I could believe that everyone would say yes. Unfortunately, there are too many people shopping around for the lowest price w/ no regard to the value add. If I had a nickel for every time that I designed a system for an engineer only to have a purchasing agent shop the parts around to find the lowest
price I could retire. I can't blame that purchasing agent because they get paid to save money. The problem I have is trying to justify
my value added resource.
Let's face it there are plenty of motion control products out there that could do 70% of all applications. How do you chose which one to use? Well if we continue down the path of motion being bought on-line, we're going to find out the hard way. Why do some manufacturers have a bad name at certain companies? Is it because they make a bad product or is it because the product was misapplied? I would say the answer is about 30/70 on that one.
Are all these organizations that are working to open up the industry really helping? ODVA has made tremendous strides to make DeviceNet a standard open field bus. That's all well and good, but all DeviceNet products are not the same. I have heard many complaints that just because it says it's DeviceNet, doesn't mean it
is. Multiple devices from multiple vendors all on the same field bus can cause multiple problems. I applaud ODVA's efforts & think
they've really helped. However, don't fool yourself into believing that there are no problems.
My point to all this isn't to prove how valuable companies like mine are. My point is to make sure that we reevaluate what we're asking for. Customers are asking for direct pricing & free support. Open architecture should not equate to openly buying these products over the internet. Quite the opposite actually; it will probably
require even more support from value added distributors because the tendency will be for manufacturers to point the finger at each other. This could change over time, but then again PC's were supposed to replace PLC's many years ago.....that's another story.
Respectfully,
Guy H. Looney
Sales Engineer
Regan Controls, Inc.
475 Metroplex Dr.
Suite 212
Nashville, TN 37211
phone: (615) 333-1940, ext. 322
fax: (615) 333-1941
[email protected]
www.regancontrols.com
>>> List Management Account <[email protected]> 03/24/00 11:41AM >>>
------- Forwarded message follows -------
From: Jon Oak <[email protected]>
Subject: MOTION: Resolver vs Encoder and Servo Drive selection
I am working on a single axis and a dual axis servo machines that are related. I am going to use a PC bus based motion card (galil, MEI, etc.) and some pc based control software. I have the option of specifying encoder or resolver but am not sure what the benefits of each technology. Being that these machine will be going into an
automtive plant environment being robust both mechanically and electically are key. Cost is also a bit of an issue, but not as important as reliability.
Question 2: Same system as above. What should you look at in a servo drive? currently I am considering the following as the requirements
Digital Drive (no pot. tuning
High voltage bus (460v Drive)
20 amp continuous 40 amps peak (min)
Communications port to talk to PC
What else should be taking into account. I tend to think of these as somewhat of a commodity item espceailly since i am using a PC motion card). I have seen a large price range $3500-2100 and
would like to make sure the lower cost options will be perform well.
I would like to use the email posted by Jon Oak (shown below) as a basis for my comments to address the "open" topic that has been so heavily discussed (John, I'll give you my 2 cents on your
application in a separate email). I have been monitoring the conversations concerning the "need" for open systems very closely & it seems to have developed into 2 distinct topics. The first is the need for open architecture and the second is the need to be able to purchase equipment directly over the internet (or from the
manufacturer) at the best price possible.
Over the past several months, many of you have used the term "vendor" with much resentment. I can only assume from the comments & wise-cracks, that many of you view vendors as worthless entities that do nothing but mark up a piece of equipment & sell it for a profit. I work for a "vendor" that considers itself a "value added distributor". I'm sure many of you will get a good chuckle out of that term & will immediately think that the only value I add is someone to beat up when your equipment is late.
I would hope that my previous posts would prove that I have some degree of competence in the motion control industry. I have only been in this industry for 4 1/2 years. However, I have done nothing but motion control for that time. I have seen hundreds of applications & feel I have a tremendous amount of experience to share in terms of what works & what doesn't. Make no mistake about it, I do not claim to be a expert & eat my share of humble pie regularly.
Jon's email proves that providing "open architecture" does not make the system any simpler to use. He's going to use a PC, some operating system (Iet the debates begin over which one he should use), a motion control card, motors, amplifiers, and some sort of mechanics. Further he could use an SST card in the PC to
plug in anybody's I/O. How much more open can you get? Okay, now we've got a truly open system and everybody's happy right? Well who's going to tell him how to make it work? Let's assume he
uses the following:
Galil's DMC-1730 motion control card w/ an Active X toolkit to interface to Visual Basic
Motors from Custom Servo Motors
Amplifiers from Pac Sci
Mechanics from THK
Gearheads from Bayside
How many of you could make this system work? This is not a straw poll, so please don't respond with "I could". The last thing I want is 300 emails telling me how to make it work. I already know how to make it work. Now for those of you that feel comfortable with the first question, how many of those people could get it up and going in less than a week? Once again no response is necessary.
My point to all of this is, that the more open you make the automation world, the more finger pointing you are going to have if you choose to deal directly with the manufacturers. Let's face it, Allen Bradley, Parker, Siemens, etc. wants you to buy their entire system. If you choose to build a system from pieces there is
going to have to be someone to step up & assist you. Integrators do not have the time to become experts on everybody's equipment. But Guy, we don't have to do that.....we'll find one system that we like & standardize on it. If you're an OEM you can do that. If you're an integrator, there's a little thing called "customer spec'd
equipment" that makes that thought process an impossibility.
It seems to me that the desire for open systems strengthens the position that "vendors" (distributors) have. Let's say I sell a
$20,000 system to do what Jon is looking for. Assume that I have a 30% GPM built in to that price. That means that if he could buy direct, he would be able to save $6,000 on capital costs. That's a lot of money right? Well let's look at what that $6,000 gets him:
Pre-sale
1) Engineering time to properly size the motors, drives, & gearheads (torque, speed, regen, inertia matching, etc.)
There are sizing software programs available to automate this task, but some are buggy & if you don't understand what is taken into account when sizing is performed you are at the mercy of the software to select the system for you. The software doesn't tell you everything that it assumes. When you give it a distance to move & a time to move it in, it probably assumes a trapezoidal move profile with a 1/3, 1/3, 1/3 time allocation for acceleration,
constant velocity, and deceleration. What it doesn't tell you is how to program the controller to achieve that move profile. If you program the wrong move profile the system may not work. Also,
what if your motor doesn't have enough speed for the move? Well you could increase the acceleration & deceleration rates to lower the top in speed requirements instead of buying a bigger motor. The opposite could be done to lower the torque & increase the top in speed. Further, many programs don't take regen into consideration or explain the importance of 2) Engineering time to make sure the system will work.
I can't tell you the number of times that I've seen gearheads break because the wrong one was chosen. People constantly try to save money by putting inferior gearheads on servos. A cheap gearhead can be used on steppers because they are either moving or not moving. Servos are always moving (even when they appear to be at rest, they are still servoing / moving). A
servo can destroy a gearhead designed for stepper systems by shaking it apart. If you're in a hot environment or moving constantly at a high rate of speed many gearheads will heat up & transfer the heat back to the motor which could damage or destroy the motor. Many gearheads have the ring gear pressed in. Further the ring gear could be made out of a different material than the housing. If that's the case, the two metals will have a different coefficient of expansion due to heat. This could lead to a lot of problems.
Orientation of the gearhead is also important. If the shaft is facing upward you will need to let the gearhead Will the motor even
mount to the mechanical system? Just because it will give enough torque and speed it doesn't mean it will physically mount. Do I need to clamp the torque of the servo to make sure I don't break
the mechanics? If so, does the control scheme (controller &/or amplifier) have that ability? Can the mechanics handle the forces & moments to give me the life I need? Do I have enough resolution in the system to ensure I can obtain the repeatability I need? Just because the mechanical system has a repeatability spec that
meets my requirements, it doesn't mean that I have enough commanded resolution to obtain that repeatability.
What about compatibility between the drives, motors, and controller? Does the inductance of the motor allow it to work w/ the amplifier? Does the switching frequency of the amplifier pose a problem to the motor? Does the controller have an encoder input great enough to handle the input frequency of the encoder? Is the motion controller's I/O compatible w/ the amplifier's I/O? In other words what if the fault output & enable input on both devices are PNP? That's gonna require extra relays to make the system work.
What about the deratings? Motor's have a torque speed curve that is based on the motor being attached to an aluminum heat sink w/ 10" x 10" x 1/4" dimensions. If you don't have that, you can burn the motor up at rated torque (I've seen it happen). Gearheads are derated based on the duty cycle and the gear ratio.
What about noise? I could write another book on that topic alone.
Post-sale
4) On-site support for hardware setup
We will usually spend at least 1 full day; many times 2 or 3 days to help a customer get up & going.
5) Phone support for programming issues
We have equipment in our office that allows us to simulate programs in the field. This allows us to debug or offer suggestions
to improve the existing program.
6) Wiring schematics to configure the entire system....enough said
7) Reduce finger pointing.
We are the ones that have to deal w/ the manufacturers pointing fingers at each other 95% of the time. I once spent 80 hours in 3 days trying to get a customer's motion controller work w/ the motors & drives I sold him. By the time it was over I knew more about the IMC-123 module than most of the AB guy's did (no, I don't sell AB products).
8) Spares on our shelf
Many manufacturers are cutting costs by becoming JIT manufacturers. This means that normally it will take anywhere from 2 days to 4 weeks to ship.
Is that worth $6,000? I wish I could believe that everyone would say yes. Unfortunately, there are too many people shopping around for the lowest price w/ no regard to the value add. If I had a nickel for every time that I designed a system for an engineer only to have a purchasing agent shop the parts around to find the lowest
price I could retire. I can't blame that purchasing agent because they get paid to save money. The problem I have is trying to justify
my value added resource.
Let's face it there are plenty of motion control products out there that could do 70% of all applications. How do you chose which one to use? Well if we continue down the path of motion being bought on-line, we're going to find out the hard way. Why do some manufacturers have a bad name at certain companies? Is it because they make a bad product or is it because the product was misapplied? I would say the answer is about 30/70 on that one.
Are all these organizations that are working to open up the industry really helping? ODVA has made tremendous strides to make DeviceNet a standard open field bus. That's all well and good, but all DeviceNet products are not the same. I have heard many complaints that just because it says it's DeviceNet, doesn't mean it
is. Multiple devices from multiple vendors all on the same field bus can cause multiple problems. I applaud ODVA's efforts & think
they've really helped. However, don't fool yourself into believing that there are no problems.
My point to all this isn't to prove how valuable companies like mine are. My point is to make sure that we reevaluate what we're asking for. Customers are asking for direct pricing & free support. Open architecture should not equate to openly buying these products over the internet. Quite the opposite actually; it will probably
require even more support from value added distributors because the tendency will be for manufacturers to point the finger at each other. This could change over time, but then again PC's were supposed to replace PLC's many years ago.....that's another story.
Respectfully,
Guy H. Looney
Sales Engineer
Regan Controls, Inc.
475 Metroplex Dr.
Suite 212
Nashville, TN 37211
phone: (615) 333-1940, ext. 322
fax: (615) 333-1941
[email protected]
www.regancontrols.com
>>> List Management Account <[email protected]> 03/24/00 11:41AM >>>
------- Forwarded message follows -------
From: Jon Oak <[email protected]>
Subject: MOTION: Resolver vs Encoder and Servo Drive selection
I am working on a single axis and a dual axis servo machines that are related. I am going to use a PC bus based motion card (galil, MEI, etc.) and some pc based control software. I have the option of specifying encoder or resolver but am not sure what the benefits of each technology. Being that these machine will be going into an
automtive plant environment being robust both mechanically and electically are key. Cost is also a bit of an issue, but not as important as reliability.
Question 2: Same system as above. What should you look at in a servo drive? currently I am considering the following as the requirements
Digital Drive (no pot. tuning
High voltage bus (460v Drive)
20 amp continuous 40 amps peak (min)
Communications port to talk to PC
What else should be taking into account. I tend to think of these as somewhat of a commodity item espceailly since i am using a PC motion card). I have seen a large price range $3500-2100 and
would like to make sure the lower cost options will be perform well.