Quote:
Originally Posted by Paul Haese
Gary to answer your question re the steel frame work. Definitely not screwed to frame, but I will check just in case. I am pretty certain it is not but will do that anyway. When I screwed the motor on I took special care not to screw to the frame. That being the case though each wall frame is isolated by a timber post in the corners and the timber floor on which it is fixed. I wonder though could this be like a faraday cage?
The USB suppressors are a good idea anyway and I will pick some up. Is this what you mean by current choke?
|
Hi Paul,
With regards the motor frame not being attached to the steel frame, very good.
Nevertheless, it still has a very strong possibility of being noise related and
the cables from the inverter to the motor are likely to be the major source of it.
What are termed "stray" capacitances which cause leakage currents can
come from a variety of sources, including from the cables themselves,
the inverter, the rectifier within the inverter and so on.
To make this concrete, these stray capacitances are created whenever two
conductors are at different potentials and are separated by a, typically thin,
insulator. That dielectric might be, for example, just plastic or even just air.
As the voltages in the motor drive cables change very rapidly, these capacitances
allow tiny currents to pass through them and their paths back to their
the source may take some large physical loop area. A conductor having a large
loop area with alternating currents, even small ones, will generate electromagnetic
noise.
In order to improve the quality of the current waveforms in motor drive systems,
high frequency switching tends to be used these days. It is the very short
rise time of the pulses of this fast switching that is one major noise culprit and
not just the pulse frequency itself. These fast rise times are harmonically rich
which means lots of noise over a wide spectrum. This multiplies the probability that
some other circuit will be susceptible to noise at one of those frequencies.
There are sources of noise and then there are victim circuits. If the ends justifies
the means, one can attempt to filter the victims. The suggestion of isolators on
your USB connections is one approach to this in that if there are common-mode
currents passing back through these cables that had originated from the motor
drive system, then isolators block this return current path.
However, my suggestion is that if indeed the problem turns out to be noise
from the motor drive, then it can be wiser to cure it at the source. This then
minimizes the chance of the problem manifesting itself in some other way,
say noise artifacts on your images.
So the common mode choke, which is a form of transformer, goes on the three
cables from your inverter to the three-phase motor. Hopefully it will "choke"
one of the largest noise contributors.
Here is an example of a class of commercial three phase filters -
http://www.digikey.com/product-highl...e-filters/1955
Here is a datasheet for one -
http://www.schaffner.com/en/products...lications.html
For belts and braces, one could add current mode choke/filter to the source and
an USB isolator on the USB cable but I would start with the choke first.
As we just discussed on the telephone, also make sure any cables running
large switching currents are not in close physical proximity to, for example, the
USB cables. Even when run at right angles to each other, there may be sufficient
coupling between them to induce noise. The USB cables should be shielded
and the design of the shield and the equipment should be such that the shields
are fully connected to the chassis at both ends, 360 degrees around.
It seems telling that when you manually switch on the motor that the problem
manifests itself. Electromagnetic interference (EMI) problems
related to inverters controlling three phase motors are common
enough that they appear in the standard texts and in the literature.
It is important to appreciate that the path of interference can be both through
conducted noise and through antenna mode radiation. In both cases, the
choke/filter can assist.
Having said all of the above, it
might be a power integrity issue with the
motor/inverter drawing large currents particularly when they first switch, but
I'd still try the common mode choke/filter first.
Best Regards
Gary Kopff
02 9457 9049
