[1000mp] 1000MP Power Supplies

Thu Sep 7 11:08:57 EDT 2006

Dear Rob,

The main root cause of failure in switching PDU's are power line transients.

Where buck'n'boost isolation xfmrs are used, like in broadband microwave
(toll) floors, the incidence is less because the windings filter out and
decrease
the voltage of the spikes, and as you know, RF has a hard time going through
xfmr
windings.

In cell sites it is absolutely horrendous because there's no isolation and
very little
filtering. Some sites have these three-phase MOV units, and that helps some,
but when your
equipment is running 24/7 on the plant it senses spikes 24/7.

We use Marconi (Lorain), Delta, C&D, Tyco, Ericsson and Purcell PDU's. The
Chinese
built stuff is the worst. I think these people are crazy, putting the
telephone plant on
Chinese rectifiers. What happens when they say "No more parts, G.I.,"? 

I see older 1231H plants that have been cranking -48 volts for over 20
years,
and most of them don't even need capacitors more than every 10 years.

Are you a Bell hand?

  Hal Mandel

-----Original Message-----
From: 1000mp-bounces at mailman.qth.net [mailto:1000mp-bounces at mailman.qth.net]
On Behalf Of Rob Atkinson
Sent: Tuesday, September 05, 2006 9:08 PM
To: All about Yaesu 1000mp
Subject: Re: [1000mp] 1000MP Power Supplies

hal,

are you aware of any consistencies in failure mode for the switching 
supplies?  be interesting and maybe useful for troubleshooting to know 
if there are one or two particular components that die, for a general 
switching supply design.

tnx

rob / k5uj

On Tuesday, September 5, 2006, at 09:37 AM, Harold Mandel wrote:

> After reading somewhere about possible switching
> noise (hash) generated by the FP-29 power supply
> it was decided to look further into the situation.
>
> Firstly, dealing with switching supplies in my
> workplace it is found that switching supplies
> have a much lower MTBF than linear supplies, and
> that we replace three times more switchers than
> linears in the telephone central office.
>
> Secondly, the issue of hash and bypassing of the FP-29
> caused me to listen to the radio and FP-29 running
> with another HF radio not connected to the system,
> but running on a battery. With no other appliances
> running there were some "birdies" in the 3 MHz
> band that were apparent and some stuff around
> 18MHz that were audible as well.
>
> So I decided to invest in a couple of Astron linear
> supplies, an RS-70M and an LS-25A.
>
> Before ever connecting these some e-mails were
> read where Astrons catch fire on the circuit board
> attaching to the one computer-grade filter cap used
> in the supplies. Both the LS and RS use only one
> capacitor, but these are different voltage ratings.
>
> The LS supply does not come with a fan. The RS
> does, and it works when a thermostat mounted on
> the right side heatsink closes. This is easily bypassed
> with a front panel switch. Both supplies grow warm
> when touching the top steel covers after an hour of
> operation.
>
> A fan from Radio Shack was added to the LS supply
> with an afternoon's work on the kitchen table. The fan
> attaches to the rear apron, mounted outside, and there
> is ample room to do so. The power is derived from
> the front panel power switch, with no thermostat.
>
> The LS is able to be adjusted beyond 30.00 volts.
> By attaching a 250 watt, 6 ohm resistor across the
> output the voltage sag observed is less than 1/2 volt
> under load from no load. The output was set at 30.00
> volts for the '1000MP.
>
> Replacement capacitors are on the way for both these units. The
> LS uses a 55 volt, 35,000uF cap. This will be replaced with
> two units of 35,000uF at 80 WVDC. Four of these caps
> were purchased, and the other two will replace the
> 35,000uF/50 volt cap in the RS unit.
>
> A Molex-type plug and/or socket was desired for hooking
> up the radio to the supply, with a jumpered arrangement
> from the other supply. No part could be found from Yaesu
> parts department. That Molex connector is not available
> at all. Another power cord was ordered so as not to destroy
> the OEM power cord for the FP-29. This was stripped and
> converted into ring terminal ends to fit the Astrons.
>
> When completed and turned up & tested, the same jig was
> used with another radio listening.
>
> Regrettably, the same "birdies" were heard. I cannot report to you
> any reduction in noise by going through this exercise.
>
> The noises heard on the '1000MP stayed the same. However, here
> in the "portable" location the RF environment is replete with all sorts
> of signals as this is near many other appliances such as computers and
> television sets, etc. It would take days to write down all the spurious
> noises heard on the HF bands. However, the "biggies" were all still 
> there.
>
> Respectfully,
>
>   Hal Mandel, W4HBM
>
>
> ______________________________________________________________
> 1000mp mailing list
> Home: http://mailman.qth.net/mailman/listinfo/1000mp
> Help: http://mailman.qth.net/mmfaq.html
> Post: mailto:1000mp at mailman.qth.net
>

______________________________________________________________
1000mp mailing list
Home: http://mailman.qth.net/mailman/listinfo/1000mp
Help: http://mailman.qth.net/mmfaq.html
Post: mailto:1000mp at mailman.qth.net