Hi. I'm new to this group and hope this is the right forum to ask this
in. I have an upcoming project that will require five VFD's to run
proportionally to a level sensor with a 4-20mA output. It will be in a
single building treating mine runoff. I'm rather new to this type of
work as my past life was one of shipboard communications. My question
is, do I just need to series the devices in the loop without concern
for how many there are? I understand the issues of interference and
wire loss and such. Is a 4-20ma signal good for that many devices?? I
appreciate any input this group may have. Thanks.
On Tue, 06 Jul 2004 17:01:03 -0400, the renowned Joe
You need to look at the voltage drop across each VFD at 20mA, sum them
up, plus the resistance of the wires * 0.02A and see if there is any
headroom left with the output compliance of the sensor (and power
supply, if it's a two-wire device).
All this information should be detailed in the specs of the devices
that you are planning to specify.
"it's the network..." "The Journey is the reward"
email@example.com Info for manufacturers: http://www.trexon.com
That is the general idea, but details count and there are several
Each load in the current loop has some resistance and that uses up
some of the available voltage driving the loop. Many drives also have
one side of their analog inputs grounded or otherwise commoned to
some voltage. This causes problems when you try to use the same
current on two or more of them. The iron clad way to share the same
input would be to add an externally powered isolator for each drive
(if the drive has a 24 volt DC output, this is a good power source).
These generally have very low load resistances (50 ohms is typical) so
3 of them in series use very little loop voltage. The isolation helps
to keep the noise generated in one drive from interfering with the
others, and if a drive fails, it can be replaced while the other two
are in service on the loop. If you mount a low voltage zener diode
across each of the isolator inputs, on a separate terminal strip
(that have a zener voltage about twice the 20 ma voltage drop of the
isolator) you can also pull and replace any one of the isolators while
the other two keep working. Having two isolators between any two
drives also limits the extent of damage during drive failure that
involves any sort of flash over or other voltage surge on the signal
I don't know if any of that is important, but it is nice to know what
the choices are.
Here is an example of the kind of isolator I am talking about (pages
Depends on the drive, many (most?) will have ground as the return for a
4-20mA input so series connection is not an option.
A simple solution may be to use 0-10v inputs on the drives in parallel with
a 500R load resistor to generate 2-10v from the 4-20mA. The drives will
probably have input offset and scale calibrations which you can adjust to
get 0-max speed from 4-20mA (if that is really what you want).
Even cheap drives have lots of fancy options. You may be able to feed
4-20mA into one drive and have it generate a 0-10v output from which you
could slave the others.
There seems to be a misunderstanding here. While it is quite possible to
series the transmitter, power supply and VFD, you are lacking a controller.
The usual arrangement is to send the level transmitter to a controller and
then the controller output to the final control element (VFD in this case).
With this arrangement you can set the desired level setpoint and adjust the
various tuning constants for optimal performance.
Without a controller your loop is the same as having a controller with a
nonadjustable setpoint of zero, a nonadjustable proportional gain of one and
no reset or derivative action at all. This MAY work in your case but is
very limited in its capabilities. To change the setpoint you have to move
the entire transmitter physically up and down. To change the effective
gain, you have to change the proportional band of the transmitter. A bad
side effect of this is that you cannot read any level outside the
proportional band unless you have two transmitters (the very old fashioned
4-20 mA is THE stand signal for all process control.
Thank you all very much for your input, it is all very helpful in
steering me in the right direction and giving me the insight to at
least ask the right questions. The simple questions often lead to
complex answers. I do like the idea of being able to replace a failed
VFD or isolator without shutting the entire system down for a small
extra bit in the planning stage. I'm afraid my superiors think it is
just a matter of running some wires. However, the signal running the
raw water pumps will no doubt need to be adjusted proportionally to
feed the pump for the polymer feed to the clarifiers and likewise for
the sludge pumps.
This group is as great as the RCM group,
but with a lot less noise ( : P
Again thanks fopr the input!
If these VFD's are driving pumps (via motors) you have to consider the
paraters of the fluids being pump(how much does the properties of
the fluids vary if any) and the characteristics of the chosen pumps. Often
VFD(motor)pump combinations will not actually move fluids until some
minimum speed is reached. As you look at your pump curves (go learn
about pump curves if this is a new concept to you) may have various
flow rates that cannot be used (such as the case when your equipment
reaches harmonic speeds and the equipment shakes excessively.
Also, you may want to ask how many pumps you will run at a give flow
rate( as 3 or 4 pumps might work for a given flow rate) depending if
you want to minimize your electricity costs, or runtime on the
equipment etc. Also, depending on your piping, valves and headers, and
automationed valves you might want alarms for such things as high
pressure, low level, low line voltage, bearing temperatures,
vibration, just to name a few. Last, as you run equipment, you will
have failures and routine maintenance. How will you automatically
manage that and provide for equipment lockout (mechanical, electrical
Best thing is to use a programmable system, like what is used
elsewhere in your facilities (SCADA, PLC, DCS, etc).
PS, keep the code (software you develop) very well documented, because
quite often, you have to go back and work on things, after some
elapsed time. The older you get, the more things and time lapses on
you..... Remembering/Relearning things when a plant/facility is under
duress, should be avoided.
Well, all of this makes very good sense and leaves me to believe that
the 'AUTO' position of the 3-pos switches we have envisioned may
seldom, if ever, get used. The system consists of a sump that collects
the raw water and is monitored by an ultrasonic level detector. The
amount of raw water entering the sump from the old mine runoff will
vary with rainfall. We plan on limiting the system to 500
gallons/hour; excess will bypass the system. The raw water from the
sump is fed via two VFD run 20 hp motor/pumps to a clarifier. The
clarifier takes the raw water and mixes it with a polymer (via 1 hp
VFD/motor/pump) to coagulate stuff in the raw water, mostly iron. The
coagulated sludge is pumped (via two 5 hp vfd/motor/xmsn/pump) to a
holding tank, and from there, floats in that tank will control feed
to the sludge press. The amount of polymer pumped to the clarifier and
the sludge pumps out of the clarifier should be in sync with the
amount of raw water being fed to the system. I guess I can't just run
some wire and hope everything works harmoniously, eh? I will need set
points for each and be able to proportion the activity of all the
VFD's so the end result is improved watershed to the Mississippi from
here in PA. There is a similar system up the road a bit, but it is all
manual and the money folks want a bit more automation. This all sure
is a change from installing and fixing satcom gear that someone else
has already designed. I'm not an engineer, just a technician. Perhaps
I have learned enough already that I can see some hired professional
help is in order...anybody out there interested???
OK, I'll bite. Before changing jobs last year, I had almost 10 years
as an EE with a company that provides control system integration
primarily for water and wastewater treatment. The outfit I work for
now is more diversified and always looking for new work. Plus,
Syracuse, NY isn't too far from Pennsylvania.
From your description, it seems like the engineering group designing
this project has covered the civil, mechanical and electrical up to
the motor control, but needs help on the details of the control
system. I have several questions:
Did you mean 500 gallons/hour or 500 gallons/minute? The latter makes
more sense with 20 HP pumps.
The polymer coagulant typically isn't just flow paced, since the amount
of stuff to be coagulated in the raw water varies. Instead, we used an
instrument called a streaming current detector to measure the surface
charge potential of the particles in the raw water downstream from the
Perhaps I am not even learning to ask the right questions yet. As I
said I am new to this field and I do not have an engineers's grasp of
the math involved in a project like this.
Yes 500 gallons per minute.
This isn't a large project, but a local solution to water quality
improvement for the sins of our forefathers.
I haven't heard anyone I work with use the term 'streaming current
detector' and as a technician, I am as close to an EE as there is in
this outfit. They are mostly chemists and mechanical types and it is a
small company. Your statement makes sense as rainfall would certainly
dilute the incoming raw water and less polymer would be needed. As an
aside, my outfit also sells chemicals so the principle of economy may
be a bit, er, conflicted? I don't really know how they think as I am a
newbie there. Are you offering service on a consultant basis? I could
set up a temp mailbox to exchange contact info if you are. I know I am
a bit over my head and may be able to contract professional input into
this project. Again thank you all for the input you have given.
On Sun, 11 Jul 2004 05:13:53 GMT, "Mike Lamond"
If by consultant you're referring to providing services to specify
the instrumentation and control requirements for the process, that's
not what we do. That would typically be done by the process engineer,
and you may or may not have a requirement for the design to be
stamped by a licensed professional engineer. We do have resources
that we can call on if needed.
What we do is system integration, where we take the engineer's spec
and provide detailed design, fabrication, programming and field
commissioning services for the control system. Installation would
be done by a local electrical contractor.
If you're interested, contact me at my work email -
lamond @ deadlinesolutions.com
Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here.
All logos and trade names are the property of their respective owners.