On Sunday 21 March 2010, Paul Keeton wrote:
>Peter,
>
> I have a question about the the 7i33. The machine we are
> discussing has three 25 HP spindle motors operating on 3 separate VFD's.
> We have a Italian control that was fitted onto a machine and it has a
> 100ma limit on its analog output. Each VFD requires a minimum of 70ma to
> run. We had to put a small phoenix contact amp on it to boost the analog
> signal to run all three drives on that machine. The problem comes when you
> turn one spindle off and only run on two. The RPM runs higher than it
> should due the decreased load on the circuit. In other words the analog
> VREF increases and the two drives have to be retuned. We get around this
> by putting a blank tool in the spindle and running all 3 wether they are
> cutting or not. Not only is this a waste of energy it also puts
> unnecessary wear on the unused spindles bearings. What is the limit of the
> analog out on the 7i33 channels? I can always use one out per spindle and
> that would completely solve the issue. What is your take on this? One
> output or three? Anyone else feel free to jump in!
>
>Paul
Paul, your choice of a signal booster appears to be a poor one. It obviously
does not have enough negative feedback to even approach the ideal 'op-amp'
characteristic that you need. While the output currents you mention are
quite high for the general description of an op-amp, many of the smaller
audio amplifier integrated circuits would have not only sufficient drive to
handle the 210+ milliamp load of all 3 spindles, but have sufficient feedback
that disconnecting and disabling 1 or 2 of the spindles would have only a .1%
effect on the remaining one.
I just dug into some parts drawers I have, and there I came across a
"TDA-1520" which is shown as a single ended device, but the input biasing is
such that it could run nicely as an op-amp from a + & - 25 volt rail supply,
and for your usage, I doubt if a + & - 15 volt (standard op-amp voltages)
supply would adversely effect it for your use.
With the example setup shown on the little simplified sheet packed into the
blister-pack, it has 30 db of gain with <.05% distortion, but a suitable
input scaling could reduce that gain of 1000 to a lessor value, and I'd
imagine additional feedback could bring it to unity gain quite nicely. You
could then put in a + or - signal of up to the std 10 volts at a small
fraction of a milliamp load on your DAC's, and track that at the output
within a small fraction of a millivolt from 0 to .5 amps of loading.
Effectively a straight piece of wire with no voltage gain but a current gain
of 100,000 or so.
There are, I am sure, at least 200 other similar, and no doubt more modern
devices being made by the chip makers which are in fact very high current op-
amps wearing an audio amplifier moniker in the chip books.
As usual, one would have to do some filtering of googles output, but there is
plenty of data out there. Almost any chip makers 'analog' or 'linear' books
should contain several useful devices, many of them available as 'samples',
aka 1 or 2 free for the asking.
>----- Original Message -----
>From: "Peter C. Wallace" <p...@mesanet.com>
>To: "Enhanced Machine Controller (EMC)" <emc-users@lists.sourceforge.net>
>Sent: Sunday, March 21, 2010 12:36 AM
>Subject: Re: [Emc-users] G52 and Fanuc conversion to EMC
>
>> On Sat, 20 Mar 2010, cogoman wrote:
>>> Date: Sat, 20 Mar 2010 16:02:39 -0400
>>> From: cogoman <cogo...@verizon.net>
>>> Reply-To: "Enhanced Machine Controller (EMC)"
>>> <emc-users@lists.sourceforge.net>
>>> To: emc-users@lists.sourceforge.net
>>> Subject: Re: [Emc-users] G52 and Fanuc conversion to EMC
>>>
>>> That price seems overkill to me too. I have been thinking about
>>> trying to develop a product for just this kind of situation. The
>>> solution seems simple, and inexpensive. Take an AVR microcontroller and
>>> have it monitor the Q-encoder signals. The program would start out
>>> counting the time between changes of the signals. When the RPMs get
>>> high enough it could switch to counting encoder events for a certain
>>> small amount of time. The RPM that is found would be converted to a DC
>>> voltage by a PWM signal from one of the counters. The PWM would always
>>> be clocked at a few MHz, so the RC filter would have a high corner
>>> frequency, and the DC voltage would be able to change faster than most
>>> (if not all) servo systems. For older servo motors that top out at
>>> about 2500 RPM this should be overkill.
>>
>> Actually theres a better way than changing counting modes at different
>> speeds
>> (from 1/T_Encoder to DeltaN/T_Sample): always timestamp encoder edges,
>> that
>> way, at T_Sample you have both encoder counts (N) and the time it took to
>> get
>> those counts (TsubN) from the timestamps. The velocity can now be
>> calculated
>> as KDeltaN/TsubN. This is what the HostMot2 hardware/driver does to
>> calculate
>> its velocity estimate (glossing over some nastiness that the driver fixes
>> at
>> reversals and stops)
>>
>> Dont know if the AVR has the hardware to do this, We've implemented this
>> on a
>> DSPIC (using DMA) but it wasn't pretty...
>>
>> Peter Wallace
>> Mesa Electronics
>>
>> (\__/)
>> (='.'=) This is Bunny. Copy and paste bunny into your
>> (")_(") signature to help him gain world domination.
>>
>>
>> -------------------------------------------------------------------------
>>----- Download Intel® Parallel Studio Eval
>> Try the new software tools for yourself. Speed compiling, find bugs
>> proactively, and fine-tune applications for parallel performance.
>> See why Intel Parallel Studio got high marks during beta.
>> http://p.sf.net/sfu/intel-sw-dev
>> _______________________________________________
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>---------------------------------------------------------------------------
>--- Download Intel® Parallel Studio Eval
>Try the new software tools for yourself. Speed compiling, find bugs
>proactively, and fine-tune applications for parallel performance.
>See why Intel Parallel Studio got high marks during beta.
>http://p.sf.net/sfu/intel-sw-dev
>_______________________________________________
>Emc-users mailing list
>Emc-users@lists.sourceforge.net
>https://lists.sourceforge.net/lists/listinfo/emc-users
>
--
Cheers, Gene
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Why use Windows, since there is a door?
(By fac...@galileo.rhein-neckar.de, Andre Fachat)
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users
