Searching \ for '[EE]:Spectrum analyzer problem, noise' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=spectrum+analyzer
Search entire site for: 'Spectrum analyzer problem, noise'.

Exact match. Not showing close matches.
PICList Thread
'[EE]:Spectrum analyzer problem, noise'
2002\09\18@094030 by Jochen Feldhaar

flavicon
face
Hi all,

I operate a spectrum analyzer model HP 853A with a 8558B RF unit, which
has a strange problem around the first LO, a YIG oscillator. As some of
you will use these also, I think it is OK to ask this question.

When switching on the unit, the noise floor on the screen is 20 dB above
the usual value, from 0 thru 700 MHz, then declining to normal at 900
MHz and higher. The effect will vanish if I tune to a CF of 1100 MHz or
higher. It will appear only if the unit has been in daily use for my
usual 8-10 hours, for several weeks continuously. Some months ago, I did
a lot of paperwork (and my vacation!) and didn't use it, the effect was
gone during the following several weeks of heavy usage, only to appear
again after some time.

I would like to know if anyone can give me a tip as to the particaular
cause of this failure mode, or perhaps point me in a direction where I
could start to dig for a new component.
Is it indeed the YIG osc, is it the first mixer or some circuitry around
this assembly (e. g. heater, driver, voltages)?

Thanks,

Jochen  Feldhaar DH6FAZ
Telejet GmbH

--
http://www.piclist.com hint: To leave the PICList
spam_OUTpiclist-unsubscribe-requestTakeThisOuTspammitvma.mit.edu


2002\09\18@115351 by Jim

flavicon
face
I can't give you any specific advice, but I would
first try to isolate the problem between the YIG osc
and it's driving circuitry.

I've got an 8558B also - and it seems to drift excessively.
I also have missing wipers off the BW switch so it is
limited to 10 and 30 KHz IF BW selections. (I added a couple
of small pieces of board foil to replace the missing
wipers.)

You know - that particular model has the first LO
routed to a BNC connector on the front panel - do
you have a second SA that you can use to observe
the spectrum of the 8558B's LO?

RF Jim


{Original Message removed}

2002\09\18@133418 by Roman Black

flavicon
face
Jochen Feldhaar wrote:

> I operate a spectrum analyzer model HP 853A with a 8558B RF unit, which
> has a strange problem around the first LO, a YIG oscillator.

> When switching on the unit, the noise floor on the screen is 20 dB above
> the usual value, from 0 thru 700 MHz, then declining to normal at 900
> MHz and higher. The effect will vanish if I tune to a CF of 1100 MHz or
> higher. It will appear only if the unit has been in daily use for my
> usual 8-10 hours, for several weeks continuously.


A common failure mode of semiconductors (especially
in small signal transistors), is a heat related fault
where after sustained heating they produce a lot of
noise. Get the unit hot with a blanket and see if the
fault occurs quicker, then try freezer spray on the
individual components in that area.

I've replaced countless small noisy transistors in
amps and older TVs, mainly appliances used for long hot
hours. And for some reason (lower Vcb ?) they are mainly
PNP small signal types. If your freezer spray points
to a component also try resoldering it first, you can
get noisy solder joints in hot equipment. And if you
trace it to a large expensive module, maybe heatsinking
or fan cooling it might save you some bucks? You do have
adequate cooling for the appliance now? ;o)
-Roman

--
http://www.piclist.com hint: To leave the PICList
.....piclist-unsubscribe-requestKILLspamspam@spam@mitvma.mit.edu


2002\09\18@154309 by Jim

flavicon
face
  "then try freezer spray on the individual
   components in that area."

Roman, in the HP 8558B the 1st LO is a YIG oscillator
in what appears to be a sealed 'can'.

Hitting it with freeze mist might have a delayed
effect ... the real problem is, without the right
mainframe or the proper extender card most of the
circuitry in these plug-ins is inaccessable.

There is the possibility that the problem is the
circuity that control the YIG osc - that's why I
suggested viewing it's RF output via the front
panel BNC jack labeled "1st LO output" - followed
up by observing (with a scope) the DC drive to the
YIG osc to nake sure that the 1st LO isn't being
'driven' funny by the driver circuit (he did
mention it reacted differently with frequency
change and this is really just a different 'tuning'
voltage applied to the YIG osc) ...

RF Jim

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestspamKILLspammitvma.mit.edu


2002\09\18@154322 by Peter L. Peres

picon face
The YIG is not a semiconductor, the transistor that drives the oscillator
is. The kind of 'high noise that ends at higher frequencies' is known to
me as a failure mode of FETs and bipolar transistors that were exposed to
ESD or reverse voltage. The fact that it comes and goes is consistent with
this but it could be a bad decoupling capacitor or something else.

Imho try to look for the fault with heat/cold. If this does not change
anything isolate the buffer and terminate its input and see what happens
to the noise. Then compare this to the inital problem.

Peter

--
http://www.piclist.com hint: To leave the PICList
.....piclist-unsubscribe-requestKILLspamspam.....mitvma.mit.edu


2002\09\19@062522 by Jochen Feldhaar

flavicon
face
Hi Jim,

that's a good tip, I will have to get my 65 Kilogram 851/8551 from home
to measure the LO.
BTW, in my unit one spring contact of the Trigger selector has fallen
off, so I'm down to video or line trigger..... HP was always the
abbreviation of high price, but in these units somehow only the price
was elevated, not the quality...... ;-(

Greets
Jochen

Jim schrieb:
{Quote hidden}

> {Original Message removed}

2002\09\19@062754 by Jochen Feldhaar

flavicon
face
Hi Roman,

see below....

Roman Black schrieb:
{Quote hidden}

It is immediately after switching on - and the memory effect needs weeks
of daily operation to make it occur, and it will even vanish after 2
months of not using the unit..... seems not like a transistor gone
bad....

Apart from that you are right.
PNP's seem to have a higher inner resistance, so your observation is
absolutely valid!

Greets Jochen
{Quote hidden}

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\09\19@081233 by Roman Black

flavicon
face
Jochen Feldhaar wrote:

> It is immediately after switching on - and the memory effect needs weeks
> of daily operation to make it occur, and it will even vanish after 2
> months of not using the unit..... seems not like a transistor gone
> bad....

Have you checked the solder joints with a high
power magnifier and bright light?
-Roman

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\09\19@085812 by Jochen Feldhaar

flavicon
face
Hi Roman,

Roman Black schrieb:
>
> Jochen Feldhaar wrote:
>
> > It is immediately after switching on - and the memory effect needs weeks
> > of daily operation to make it occur, and it will even vanish after 2
> > months of not using the unit..... seems not like a transistor gone
> > bad....
>
> Have you checked the solder joints with a high
> power magnifier and bright light?
Roman, this is not a cheap japanese CD player - there is almost 1 cm of
polished brass in every direction, and all cables are soldered very
carefully indeed. But yes, I have tried by tweaking, looking, gently
(and not so gently) tapping on components and so on - no effect!
And it goes away when I tune to a different Center Frequency - and in a
YIG this is done with a DC current, but as I am already scanning a 1 GHz
window and only altering the CF, and the mistake appears in EVERY sweep,
but only on the left side, I don't think it can be a contact - must be
something Mr. Murphy invented exclusively for my personal well-being....

Greets Jochen

> -Roman
>
> --
> http://www.piclist.com hint: PICList Posts must start with ONE topic:
> [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


More... (looser matching)
- Last day of these posts
- In 2002 , 2003 only
- Today
- New search...