Robert Chudek - K0RC wrote:
..."On the other hand, I would expect the Factory to investigate this
issue. For example, they could throw a 1,000 feet of cable between a
controller and 4 EHU's and get their scope out. Then take a look at the
pulse train at the Power Supply/Controller end and compare it at each
EHU. This puts the engineering back where it belongs, at the Factory. "...
Best suggestion I have heard lately. I certainly wouldn't recommend the
user go change his cable when the factory doesn't know the maximum
limits for the cable.
Jerry
Robert Chudek - K0RC wrote:
>Here, pass that steering wheel into the backseat where I'm sitting... I might
>as well give this a shot too... :-)
>
>I have been very interested in reading this thread and the good suggestions of
>cause, effect, remedy. I think the suggestion from Bill - W2WO is a practical
>test to divide and conquer the problem. This procedure requires the customer
>to do some Field Engineering,. In my opinion, a "service junction box" should
>be installed at the base of the tower in any case.
>
>On the other hand, I would expect the Factory to investigate this issue. For
>example, they could throw a 1,000 feet of cable between a controller and 4
>EHU's and get their scope out. Then take a look at the pulse train at the
>Power Supply/Controller end and compare it at each EHU. This puts the
>engineering back where it belongs, at the Factory. (I suspect pulse
>deterioration from one end of the cable to the other, as has been suggested.)
>
>The next Factory step would be to cut the cable in half and compare the
>results again. An approximation of cable length effect could then be
>determined with the expectation of determining a "maximum cable length"
>specification. Ultimately, this testing could identify what controller cable
>type would be acceptable with very long runs (if the cable IS the issue).
>
>Using the "shotgun" replacement approach is certainly acceptable in normal
>cases. (It was faster and more economical to replace computer boards than
>troubleshoot to the component level in the field.) This approach gets the
>client up and running promptly. But in this particular situation, a 500 foot
>controller line doesn't seem to fall into the "normal" installation to me.
>
>So that's 4¢ from me (inflation), and I'm anxiously waiting to hear about the
>final remedy!
>
>73 de Bob - KØRC in MN
>
>(Retired Field Service Engineer - Linotype/Hell Company)
>
>
>------------------------------
>
>Message: 4
>Date: Mon, 11 Jun 2007 10:51:10 -0400
>From: Bill Ogden <ogden@us.ibm.com>
>Subject: [TowerTalk] SteppIR problem
>To: towertalk@contesting.com
>Message-ID:
><OFEC916277.AE67A773-ON852572F7.00546E0C-852572F7.0054F698@us.ibm.com>
>Content-Type: text/plain; charset=US-ASCII
>
>Simple suggestion: Ask the SteppIR people for a control-box connector with
>a few feet of cable on it. (That is, do not try to solder one yourself. It
>is difficult to do accurately without a proper jig.) Take this to the base
>of the tower. Cut the existing cable and connect all the pigtale lines at
>this point. A little messy with the 16+ wires, but it should be safe and
>accurate. Operating the control box from this point should resolve all the
>questions about the 500' cable. Of course, this assumes you can get 110v
>power to the tower base. Consider borrowing a generator, if necessary.
>
>Bill - W2WO
>_______________________________________________
>
>
>
>_______________________________________________
>TowerTalk mailing list
>TowerTalk@contesting.com
>http://lists.contesting.com/mailman/listinfo/towertalk
>
>
>
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|