Re: Agilent E4404B Spectrum Analyzer With Dead PSU Repair
Pittsburgh, not much lightning here.
Nearly every Rifa capacitor I see is swollen and crazed, and when you see that, detonation isn't far behind.
-Dave
toggle quoted message
Show quoted text
On 4/22/24 17:55, Jim Ford wrote: I suspect you live in lightning country, Dave.? Southern California where I live is just about the least likely place on the planet for lightning strikes.? After damaging at least 2 instruments in the process of replacing (old) RIFAs, I'm not risking that again.? YMMV.
Jim
------ Original Message ------
From "Dave McGuire" <mcguire@...>
To [email protected]
Date 4/22/2024 4:12:11 PM
Subject Re: [HP-Agilent-Keysight-equipment] Agilent E4404B Spectrum Analyzer With Dead PSU Repair
On 4/22/24 14:07, Jim Ford wrote:
But I don't replace old RIFAs on sight anymore.? Too much collateral damage by a ham-fisted engineer with a soldering iron in his hand, and explosions are rare here in 120 VAC land.? Those in 240 VAC land risk explosions, though. ? Uh, I've had three go so far just this year.? The risk is surely higher in 240V-land, but I strongly advise replacing them on sight here too.
?????????? -Dave
-- Dave McGuire, AK4HZ
New Kensington, PA
--
Dave McGuire, AK4HZ
New Kensington, PA
|
It isn¡¯t too much to ask for that you be able to replace commodity components that fail with identical replacements. It¡¯s also not unreasonable to want to replace a board with a functional equivalent with the understanding that the original manufacturer might choose not to provide tech support (free, paid by subscription, or pay per issue).
In the computer industry, the Moss Magnuson Act (US) doesn¡¯t allow a manufacturer to void a warranty for using compatible aftermarket parts, like commodity DRAM modules or fiber optic modules. They¡¯re not required to support or troubleshoot compatibility issues. You could end up having to replace the commodity components if you end up at an impasse obtaining support under a contract (and then swapping back after problem resolution). A common example would be a cisco router with Cisco charging stupid amounts of money for DRAM modules or fiber transeivers.
The embedded PC motherboards are specialized enough that it would potentially be difficult but not impossible to find a replacement. I don¡¯t know that it¡¯s come up yet, but I¡¯m other countries like the UK, where a manufacturer can¡¯t tie a license to a piece of hardware¡in principle someone might be able to turn the screws and successfully litigate their way into forcing a manufacturer to make a license portable, or by extension license a replacement board.
If only I had a UK mailing address and corporate entity¡
|
Re: Agilent E4404B Spectrum Analyzer With Dead PSU Repair
I suspect you live in lightning country, Dave. Southern California where I live is just about the least likely place on the planet for lightning strikes. After damaging at least 2 instruments in the process of replacing (old) RIFAs, I'm not risking that again. YMMV.
Jim
toggle quoted message
Show quoted text
------ Original Message ------ From "Dave McGuire" <mcguire@...> To [email protected]Date 4/22/2024 4:12:11 PM Subject Re: [HP-Agilent-Keysight-equipment] Agilent E4404B Spectrum Analyzer With Dead PSU Repair On 4/22/24 14:07, Jim Ford wrote:
But I don't replace old RIFAs on sight anymore. Too much collateral damage by a ham-fisted engineer with a soldering iron in his hand, and explosions are rare here in 120 VAC land. Those in 240 VAC land risk explosions, though. Uh, I've had three go so far just this year. The risk is surely higher in 240V-land, but I strongly advise replacing them on sight here too.
-Dave
-- Dave McGuire, AK4HZ
New Kensington, PA
|
Ah, the E4406A is the one with the hole around 321.4 MHz, isn't it?? That could be really annoying if you were working with say, 225-450 MHz, as I was a few years back.? I do have it on my wishlist, nonetheless, as they were going for only a few hundred dollars when the wireless carriers started dumping them.
Jim Ford Laguna Hills, California, USA
"And we know that God causes all things to work together for the good of those who love God and are called according to his purpose for them." Romans 8:28 (NLT)
------ Original Message ------
Date 4/22/2024 6:43:00 PM
Subject Re: [HP-Agilent-Keysight-equipment] N9020A
If it helps promote how good these MXA analysers are, there is a nice hardware feature in the IF section of the PSA and the PXA and MXA analysers that helps when making critical dynamic range measurements.
There should be a suite of IF preselector filters that limit the bandwidth that the digital IF is exposed to. On the PSA, this means there is a lumped or crystal filter that gets selected to provide this preselection automatically based on span/RBW settings. I don't know how many IF preselector filters there are in the PSA but there are also a few preselector filters like this in the PXA and MXA. The IF bandwidth can be adjusted from a few kHz up to about 1.8MHz or so by selecting a suitable preselector.
This means that these analysers can still compete with the traditional swept analysers that use analogue RBW filters (like the HP 8568B) when trying to detect very small signals near very large ones because the ADC can be protected from the nearby large signal by the preselector. Lower cost instruments with a digital IF won't have this suite of preselectors and the ADC can easily be overloaded by signals many MHz away.
The old Agilent E4406A is a bit of an oddball because it has a very crude frequency plan that compromises the image and IF rejection. It was never meant to be used as a spectrum analyser and was meant to be used under controlled conditions to look at commercial transmitters used for mobile phones etc. I think that's why it says 'Transmitter Tester' on the front panel. This was probably Agilent's way of justifying the oddball frequency plan it uses! The E4406A should have a similar suite of preselector filters ahead of the digital IF but the E4406A only operates in FFT mode (not swept) so the benefit isn't as great.
I've still got my E4406A and still use it now and again. I bought the version with the 14 bit digital IF which was important to me at the time.
|
Re: HP 140B LVPS -12.6V Not Working
I've been cross-checking the board against the schematic.? I thought I had the dot convention correct (emitter) but then the differences between a few of the transistors on the board made me realize something didn't match so I need to be even more careful if I end up needing to replace those.
I'm just glad wires don't have dots.
Thanks,
Barry - N4BUQ
I would've guessed that the dot marks the collector. At least that
was the convention used on all the parts I encountered as a kid. The
literature of the day often referred to "the collector dot". But of
course that doesn't mean that other companies/countries followed
that convention, so I'm curious: Are you just guessing (say, based
on a quick DMM diode check, in which case you're rolling the dice),
or did you cross-check with some circuit-tracing against a
schematic?
And if you're talking about a TO-92, there is no standard pinout.
Parts of US design generally follow EBC, UK parts often follow CBE
(making it easy to swap with US parts by simply installing the other
way around), Japanese parts are frequently BEC (I've also
encountered ECB), but "generally" does mean "not always" here. You
have to check, don't assume.
Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
420 Via Palou Mall
Stanford University
Stanford, CA 94305-4070
On 4/22/2024 3:38 PM, n4buq wrote:
Something about this is confusing
me.? Looking at the board, the transistors and diodes all have
a dot at one lead.? I was thinking that dot indicated the
emitter and for some transistors, that holds.? It agrees with
the 2n3409 I replaced and most of the metal-hat transistors.?
With the flat of the 2n3904's TO92 facing me, left to right is
EBC.? For Q484, though, the diode across BE (I can't read it
very well but I think that's CR485 or CR486) is connected to
the center lead and the right-hand lead (facing the
transistor's flat).? Do those not follow the same convention
as, say, a 2n3904?? I'm used to either EBC being left-to-right
facing the flat or, possibly, B on the left or right, but this
is puzzling to me.
|
If it helps promote how good these MXA analysers are, there is a nice hardware feature in the IF section of the PSA and the PXA and MXA analysers that helps when making critical dynamic range measurements.
There should be a suite of IF preselector filters that limit the bandwidth that the digital IF is exposed to. On the PSA, this means there is a lumped or crystal filter that gets selected to provide this preselection automatically based on span/RBW settings. I don't know how many IF preselector filters there are in the PSA but there are also a few preselector filters like this in the PXA and MXA. The IF bandwidth can be adjusted from a few kHz up to about 1.8MHz or so by selecting a suitable preselector.
This means that these analysers can still compete with the traditional swept analysers that use analogue RBW filters (like the HP 8568B) when trying to detect very small signals near very large ones because the ADC can be protected from the nearby large signal by the preselector. Lower cost instruments with a digital IF won't have this suite of preselectors and the ADC can easily be overloaded by signals many MHz away.
The old Agilent E4406A is a bit of an oddball because it has a very crude frequency plan that compromises the image and IF rejection. It was never meant to be used as a spectrum analyser and was meant to be used under controlled conditions to look at commercial transmitters used for mobile phones etc. I think that's why it says 'Transmitter Tester' on the front panel. This was probably Agilent's way of justifying the oddball frequency plan it uses! The E4406A should have a similar suite of preselector filters ahead of the digital IF but the E4406A only operates in FFT mode (not swept) so the benefit isn't as great.
I've still got my E4406A and still use it now and again. I bought the version with the 14 bit digital IF which was important to me at the time.
|
! :)
--
Prof. Thomas H. Lee
Faculty Co-Director, SystemX Alliance
Director, Stanford-Samsung Research Initiative
Allen Ctr., Rm. 205
420 Via Palou Mall
Stanford University
Stanford, CA 94305-4070
toggle quoted message
Show quoted text
On 4/22/2024 4:21 PM, Peter Gottlieb wrote: That's known as improving your image!
|
That's known as improving your image!
toggle quoted message
Show quoted text
On 4/22/2024 7:19 PM, Tom Lee via groups.io wrote: That's almost image amplification!
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
420 Via Palou Mall
Stanford University
Stanford, CA 94305-4070
On 4/22/2024 4:15 PM, John Miles wrote:
I¡¯d hate to get by without the E4406A, that¡¯s for sure.? It sees a lot of use these days looking at L-band signals.
But speaking of image rejection¡ LOL! You get 60-80 dB with the MXA, 45-55 dB with the SM435C, and a whopping 0 dB with the E4406A. ?
-- john, KE5FX
|
That's almost image amplification!
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
420 Via Palou Mall
Stanford University
Stanford, CA 94305-4070
On 4/22/2024 4:15 PM, John Miles wrote:
toggle quoted message
Show quoted text
I¡¯d hate to get by without the E4406A,
that¡¯s for sure.? It sees a lot of use these days looking at
L-band signals.?
?
But speaking of image rejection¡ LOL!? You
get 60-80 dB with the MXA, 45-55 dB with the SM435C, and a
whopping 0 dB with the E4406A. ?
?
-- john, KE5FX
|
I¡¯d hate to get by without the E4406A, that¡¯s for sure.? It sees a lot of use these days looking at L-band signals.? ? But speaking of image rejection¡ LOL!? You get 60-80 dB with the MXA, 45-55 dB with the SM435C, and a whopping 0 dB with the E4406A. ? ? -- john, KE5FX ?
toggle quoted message
Show quoted text
From: [email protected] < [email protected]> On Behalf Of hardyhansendk via groups.io Sent: Monday, April 22, 2024 1:07 PM To: [email protected]Subject: Re: [HP-Agilent-Keysight-equipment] N9020A ? Hi John You are right¡the top notch versions from signal hound are very impressive ¨CI were referring to their models matching the price---below 2k¡all have a price .As jeremy points to--- we just experience ?a great new ??deal as we had with E4406A (recall you bought two-- Still in use? )The ?MXA are not the best ¡ but great value below 2k¡¯ Hardy
|
Re: Is it worth looking for a R/3*2 Instrument Controller? - Feedback request
Hi Tony:
When I was working many decades ago the cost of any of the HP RMB controllers was very high.? Then Sun came out with a diskless node workstation.
HP responded with the 318 Diskless Unix Workstation and I ended up buying a number of them for use a RMB instrument controllers.? These cost way less.
--
Have Fun,
Brooke Clarke
axioms:
1. The extent to which you can fix or improve something will be limited by how well you understand how it works.
2. Everybody, with no exceptions, holds false beliefs.
|
Re: HP 140B LVPS -12.6V Not Working
I would've guessed that the dot marks the collector. At least that
was the convention used on all the parts I encountered as a kid. The
literature of the day often referred to "the collector dot". But of
course that doesn't mean that other companies/countries followed
that convention, so I'm curious: Are you just guessing (say, based
on a quick DMM diode check, in which case you're rolling the dice),
or did you cross-check with some circuit-tracing against a
schematic?
And if you're talking about a TO-92, there is no standard pinout.
Parts of US design generally follow EBC, UK parts often follow CBE
(making it easy to swap with US parts by simply installing the other
way around), Japanese parts are frequently BEC (I've also
encountered ECB), but "generally" does mean "not always" here. You
have to check, don't assume.
Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
420 Via Palou Mall
Stanford University
Stanford, CA 94305-4070
On 4/22/2024 3:38 PM, n4buq wrote:
toggle quoted message
Show quoted text
Something about this is confusing
me.? Looking at the board, the transistors and diodes all have
a dot at one lead.? I was thinking that dot indicated the
emitter and for some transistors, that holds.? It agrees with
the 2n3409 I replaced and most of the metal-hat transistors.?
With the flat of the 2n3904's TO92 facing me, left to right is
EBC.? For Q484, though, the diode across BE (I can't read it
very well but I think that's CR485 or CR486) is connected to
the center lead and the right-hand lead (facing the
transistor's flat).? Do those not follow the same convention
as, say, a 2n3904?? I'm used to either EBC being left-to-right
facing the flat or, possibly, B on the left or right, but this
is puzzling to me.
|
Re: HP 140B LVPS -12.6V Not Working
The problem is there is no -12.6V.? That sits a zero and I can't determine why.? There is sufficient voltage at the collector of Q481.? If Q481 were open that could cause it.? If -12.6V was overloaded, then Q483 should be shutting down Q481 but I can't see where either is the case. I've checked the resistors - at least in circuit - and I don't find anything wrong there but until I can get something going for -12.6V, nothing is checking out correctly.
Thanks, Barry - N4BUQ
You should have more than 0.7V across BC of Q484 in normal operation, E is connected to the -12.6V output, B should be approx -12V and C +2V.
Another common problem is some of the larger film resistors, they either drift up in value or go open circuit, this is a problem in the 140, 141, 180, 181 etc series. R485 & R489 are probably most likely to suffer from this in the circuit part you posted, as they have 90+ volts to drop. Had this with that 180A in the horizontal section, both 22.1k resistors had been cooked and gone up in value.
David
|
Re: HP 140B LVPS -12.6V Not Working
Something about this is confusing me.? Looking at the board, the transistors and diodes all have a dot at one lead.? I was thinking that dot indicated the emitter and for some transistors, that holds.? It agrees with the 2n3409 I replaced and most of the metal-hat transistors.? With the flat of the 2n3904's TO92 facing me, left to right is EBC.? For Q484, though, the diode across BE (I can't read it very well but I think that's CR485 or CR486) is connected to the center lead and the right-hand lead (facing the transistor's flat).? Do those not follow the same convention as, say, a 2n3904?? I'm used to either EBC being left-to-right facing the flat or, possibly, B on the left or right, but this is puzzling to me.
Thanks again, Barry - N4BUQ
With the different variants of the 4-071 you need to make sure the leads are in the right configuration, one of the old 4-071 variants has leads formed to get the base in the middle. TI is Texas Instruments, the 001 is a date/batch code, the Motorola ones usually have EBC marked on them.
David
|
Is it worth looking for a R/3*2 Instrument Controller? - Feedback request
Hey All,
I've always wondered if I should try to acquire one of the R/3*2 Instrument Controllers (332, 362, 382)
If anyone has one, would you be able to provide some feedback on how you use it and why you keep it? I've always liked the idea of having one in my rack and I've just started working with the hobbyist license for HT BASIC so I can get a handle on what using the original HP BASIC would be like but I'm just wondering if it is worth trying to acquire and piece together one of these seeing that the ones I can find today are all broken or missing things like keyboards etc or if I should just give it a pass and keep working on HT BASIC for Windows.
Would love to hear your thoughts.
TonyG
|
Re: HP 140B LVPS -12.6V Not Working
You should have more than 0.7V across BC of Q484 in normal operation, E is connected to the -12.6V output, B should be approx -12V and C +2V.
Another common problem is some of the larger film resistors, they either drift up in value or go open circuit, this is a problem in the 140, 141, 180, 181 etc series. R485 & R489 are probably most likely to suffer from this in the circuit part you posted, as they have 90+ volts to drop. Had this with that 180A in the horizontal section, both 22.1k resistors had been cooked and gone up in value.
David
|
Re: Agilent E4404B Spectrum Analyzer With Dead PSU Repair
On 4/22/24 14:07, Jim Ford wrote: But I don't replace old RIFAs on sight anymore.? Too much collateral damage by a ham-fisted engineer with a soldering iron in his hand, and explosions are rare here in 120 VAC land.? Those in 240 VAC land risk explosions, though. Uh, I've had three go so far just this year. The risk is surely higher in 240V-land, but I strongly advise replacing them on sight here too. -Dave -- Dave McGuire, AK4HZ New Kensington, PA
|
On 4/22/24 14:51, Peter Gottlieb wrote: So what options should I look for in one of these MXA units? I'm not sure yet. I asked from an ebay seller and got out of focus photos:
|
Re: Agilent E4404B Spectrum Analyzer With Dead PSU Repair
having had quite a few
go here in 120 VAC land...if I am in the equipment I replace (or at least remove )
especially if easy access and they are yellow before the stench
and mess....I cannot stand the stench or the mess which is the
worst part. btw the stench is not so bad in summer or when
weather allows windows and doors open...winter is another
matter...and when do they like to fail.......
¸é±ð²Ô¨¦±ð
Northern California Mtns
On 4/22/24 11:07 AM, Jim Ford via
groups.io wrote:
toggle quoted message
Show quoted text
Yes, I've bought new RIFA caps to replace the old ones.? WIMA
and Kemet are two other brands that make the safety rated X and
Y caps.
But I don't replace old RIFAs on sight anymore.? Too much
collateral damage by a ham-fisted engineer with a soldering iron
in his hand, and explosions are rare here in 120 VAC land.?
Those in 240 VAC land risk explosions, though.
HTH.
Jim Ford
Laguna Hills, California, USA
"And we know that God causes all things to work together
for the good of those who love God and are called according
to his purpose for them." Romans 8:28 (NLT)
------ Original Message ------
Date 4/22/2024 2:27:58 PM
Subject Re: [HP-Agilent-Keysight-equipment] Agilent E4404B
Spectrum Analyzer With Dead PSU Repair
On 4/22/24 15:22, Allen Hill via
groups.io wrote:
Are RIF caps still used in new
equipment, or have manufacturers found better solutions?
?
The common RIFA capacitor failure is
a materials problem, which was solved long ago. Later RIFA
capacitors do not exhibit the explosion problem. Many
companies manufacture safety capacitors.
?
-Dave
?
-- Dave McGuire, AK4HZ
New Kensington, PA
?
?
?
?
?
|
So what options should I look for in one of these MXA units?
There are so many posts in the thread I can't find the one with the link to the well-optioned units.
Peter
toggle quoted message
Show quoted text
On 4/22/2024 4:48 PM, John Griessen via groups.io wrote: On 4/21/24 22:44, Tom via Metronet wrote:
last I heard, an original X-series can be upgraded to the lates Wintel version just by replacing that module. The RF hardware doesn¡¯t need to change at all. Some of the aliexpress sellers offer a bare metal MXA without all the options for $150.? It could be a parts donor to upgrade an old MXA, but then how to add options?
|
Dave McGuire
factory
