Re: HP Semiconductor Analyzers
"Jared Cabot via groups.io" <jaredcabot@...> writes: On Fri, Sep 9, 2022 at 05:40 AM, Sergey Kubushyn wrote:
Most common failure in 1345 is one (or more) of the trimpots lost
connection in one of terminals. Those small Bourns round trimpots are a real
disaster. I had at least a couple of dozens of those replaced in all kinds of
HP and Tektronix gear and that was the only parts failed. That included 3 or
so 1345 displays.
I found the same in the 1345 displays in both my 4145A and 4145B, more
than one of those little Bourns 3329-series
trimpots were problematic. The interesting thing was that it wasn't even noticable when turning the pot. It was just open when the unit was cold, and after it warmed up, everything was fine. I just suspected the pot because during after some warmup time the width of the picture changed in a 'flickering' way until it finally worked ok.
|
Read frequency list from HP 8510C via HP-IB
Hi
I'm trying to read the list of stepped frequency points for the current
(corrected) trace on an HP 8510C via HP-IB. I *think* OUTPFREL is
supposed to accomplish this, but I see the message:
CAUTION: REQUESTED DATA NOT CURRENTLY AVAILABLE
The manual says this about that error message:
You tried to input/output raw data arrays that were not available due to
the calibration type, or tried to input calibration coefficients without
selecting the proper calibration type (see INPURAW, OUTPRAW), or tried
to output a memory trace without turning on memory.
Only the last option there seems like it could be applicable. What does
"tried to output a memory trace without turning on memory" mean?
FYI I'm able to read out the corrected trace data just fine with
OUTPDATA (I've set the format to ASCII = FORM4).
How can I read out the frequency stimulus points?
Thanks in advance,
Matt
|
Re: HP 3580A Spectrum Analyzer - broken frequency display
Hi Marco,
The scope bezel does come off, but because of age the plastic will have become brittle, so you do have to be careful. There are two tabs on the bezel which lock it into a metal surround on the front panel.?
These tabs are hidden but are located, one on the top and the other on the bottom in the center of the bezel. If you use your thumb and middle finger of one hand to grip the top corners of the bezel, ready to pull it forward when the top tab is released, then with your other thumb press down on the center of the bezel near the rear edge. This should release the top part of the bezel from the instrument.
As for the board it is small enough to be sent in a letter so the postage should not be expensive. This is just a blank board, no parts included. The seven segment displays I used are available from RS Components, E14/Farnell, Digikey, and Mouser.
As for the right-angle header the pins are longer those currently available, so I just reused the one from the original display board.
Let me know you mailing address and I'll send the board.
Regards,
Peter
|
70427A, or 70428A, or 71707A, or 717108A, or N5507A service information?
Does anyone have access to service information and/or schematics for any of these microwave sources and down-converters?
As far as I can tell HP/Agilent never released any service info on any of these fine instruments. Thank you.
|
Re: HP Semiconductor Analyzers
On Fri, Sep 9, 2022 at 05:40 AM, Sergey Kubushyn wrote:
Most common failure in 1345 is one (or more) of the trimpots lost connection
in one of terminals. Those small Bourns round trimpots are a real disaster.
I had at least a couple of dozens of those replaced in all kinds of HP and
Tektronix gear and that was the only parts failed. That included 3 or so
1345 displays.
---
*
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
*
I found the same in the 1345 displays in both my 4145A and 4145B, more than one of those little Bourns 3329-series trimpots were problematic. Jared
|
Re: HP Semiconductor Analyzers
I have a spare boot floppy for the 4145A, can make a copy on a new floppy disk for the 4145B, and have a set of Agilent floppies in a sealed box for the 4155B/4156B (Kit# 04155-85022, containing P/N: 04155-17001, 04155-17004, 04155-17401, and E5250-17001)
I'm able to take images of these disks if you like (not the 4145A disk yet, still working on getting the 5.25 floppy drive operational) and email them, or sell the physical disks for the cost of shipping plus what I paid for them (a nominal amount).
Regards,
Jared
|
Re: HP Semiconductor Analyzers
On Thu, 8 Sep 2022, Sven Schnelle wrote: "Sven Schnelle" <svens@...> writes:
The HP4145A + B are likely easy repairs, on one the 1345 XY display
sometimes only shows a vertical line, pointing to a problem in the X
deflection. The X/Y outputs at the back of the unit are showing the
proper signal. The tube isn't getting any signal, so it must be the
output amplifier. Might be just the gain pot - it says 1k, but i measure 1.6k. I put a
fixed 270Ohm resistor now because i don't have the right pot, lets see
whether that helps. Crossing fingers... Most common failure in 1345 is one (or more) of the trimpots lost connection in one of terminals. Those small Bourns round trimpots are a real disaster. I had at least a couple of dozens of those replaced in all kinds of HP and Tektronix gear and that was the only parts failed. That included 3 or so 1345 displays. --- * * KSI@home KOI8 Net < > The impossible we do immediately. * * Las Vegas NV, USA < > Miracles require 24-hour notice. * *
|
Re: HP Semiconductor Analyzers
"Sven Schnelle" <svens@...> writes: The HP4145A + B are likely easy repairs, on one the 1345 XY display
sometimes only shows a vertical line, pointing to a problem in the X
deflection. The X/Y outputs at the back of the unit are showing the
proper signal. The tube isn't getting any signal, so it must be the
output amplifier. Might be just the gain pot - it says 1k, but i measure 1.6k. I put a fixed 270Ohm resistor now because i don't have the right pot, lets see whether that helps. Crossing fingers... /Sven
|
HP Semiconductor Analyzers
Hi List,
i couldn't resist and bought a lot of HP Semiconductor Analyzers.
In particular:
HP4145A
HP4145B
HP4155B
HP4156A
The HP4145A + B are likely easy repairs, on one the 1345 XY display
sometimes only shows a vertical line, pointing to a problem in the X
deflection. The X/Y outputs at the back of the unit are showing the
proper signal. The tube isn't getting any signal, so it must be the
output amplifier.
The other one is missing the boot floppy. Everything else seems to work,
i confirmed this by taking the HxC emulator which is already present in
the other unit to boot. So i'll likely remove the floppy drive and just
put in a HxC emulator and be done. Although i don't need two analyzers,
i mainly bought them to make sure i have one if i need some unobtainable
parts.
The HP4155/56 seem to be a bit more problematic - both have broken power
supplies. One is not starting at all, the other one is starting for a
second and shuts down afterwards.
I haven't found a schematic/service guide for them - does anyone know
any details about the power supplies? One is a Voltex D2160, the other
one i don't know yet. But they look quite sophisticated, with a lot of
different voltages and several PCBs.
Thanks
Sven
|
"Sven Schnelle" <svens@...> writes: So if the YTM current is not parked correctly the YIG sphere will interfere with the Band 0 passband. This should be
fairly straightforward to check by measuring the coil voltage around 4GHz and confirming it parks at a similar location
when a Band 0 frequency is selected. I think i measured ~2.3V above the coil, but don't know what the voltage
required to tune it to 4GHz is. I'll check this evening, thanks!
A mismatched YTF interfering would also explain why the dip is at ~2.2Ghz,
and after that the output level increases again...
I did another test now: unsolder the tuning coils from the PCB and connect it to voltage source. Moving current from 0 to 50mA (i think i've read 15mA/GHz in the Service manual) didn't change the level of the output. I've found the SYTM now used and ordered two. Lets see whether at least one of them works. If it works, i'm selling the other one - guess i'll find someone who needs one. /Sven
|
"Bruce" <bruce@...> writes: I'd toa bit more investigation:
1) Test YIG output (rar of 8340)
2) Test input to the YTM from the low band amp
Things to suspect: Low band AMP, heterodyne mixer and 3.8GHz oscillator.
SRD is not involved in band 0 - PIN switch is involved in band 0 I think i checked both a while ago, but will check again this evening. Thanks!
|
Hi Wayne, "Wayne ZL2BKC" <w.knowles@...> writes: Sven,
The graph you included showing the insertion loss for band 0 now has me thinking. The slope you measure is almost
identical to what I witness against my faulty unit (with a falloff from ~1 to ~2.3GHz).
Digging into the 8340A service manual page 8-827 states:
When the instrument is in the Band 0 the SYTM magnet is tuned to about 4 GHz to keep the YIG sphere from
interferring with the output. |Tuning the SYTM to 4 GHz is accomplished by closing Q17. R109, Q21, | and Q25 will be
discussed in the Block J description (kick pulses).
So if the YTM current is not parked correctly the YIG sphere will interfere with the Band 0 passband. This should be
fairly straightforward to check by measuring the coil voltage around 4GHz and confirming it parks at a similar location
when a Band 0 frequency is selected. I think i measured ~2.3V above the coil, but don't know what the voltage required to tune it to 4GHz is. I'll check this evening, thanks! A mismatched YTF interfering would also explain why the dip is at ~2.2Ghz, and after that the output level increases again... Thanks Sven
|
Re: HP 3580A Spectrum Analyzer - broken frequency display
Hi Peter,
I humbly apologize for my silence but I had no time to attend the group and accordingly, I completely missed your message.
Thank you so much for your offer!!
I live in Italy: do you think you can manage shipping (cost on my side of course)?
I over and over saw your video and I'd like to ask you how you managed to draw out the frequency meter block.
Il looks stuck by the scope bezel I can't remove.
I'm afraid to crack something applying force.
Thank you
|
...and the discussion gets just about as heated! On 08/09/2022 00:44, Dr. David Kirkby, Kirkby Microwave Ltd wrote: As much as I prefer bottom-posting on public mailing lists, arguing about the merits of one way or the other is as pointless as arguing about religion.
oops, I mean ...and the discussion gets just about as heated!
|
Re: Hp 8590a eeprom problem.
On Thu, Sep 8, 2022 at 03:42 AM, Richard Cook wrote:
t's indicating eeprom u21
What others have said. Please check the type of EEPROM. They can be programmed in different ways.
|
Hi all,
Thanks again for great advice on my thread about spectrum analyzers.
I'm looking to expand my capabilities with one or more signal generators. I'm mostly interested in pulse modulation and highly accurate control of pulse timing and signal phase for experiments in the MIMO/phased array space.
I have been thinking about picking up an 8341B and/or an 8673B. Anyone who wants to share an opinion is welcome! If you have different ideas about what equipment I should be consifering, I'd love to hear it.
I will start somewhere in the S-K bands but I'd like to work my way all the way up to around 100GHz.
I've also seen screw-type waveguide phase adjusters that I believe will be useful as I push into the mmwave range, but I don't know a lot about what to look for.
Good support in HP-IB and Linux GPIB specifically is a plus since that should help me automate my tests.
I believe I will also need to consider an external clock source, so advice about that is also welcome!
Thank you in advance!
|
Sven, The graph you included showing the insertion loss for band 0 now has me thinking.?? The slope you measure is almost identical to what I witness against my faulty unit (with a falloff from ~1 to ~2.3GHz). Digging into the 8340A service manual page 8-827 states:
?When the instrument is in the Band 0 the SYTM magnet is tuned to about 4 GHz to keep the YIG sphere from interferring with the output. |Tuning the SYTM to 4 GHz is accomplished by closing Q17. R109, Q21, |? and Q25 will be discussed in the Block J description (kick pulses).
So if the YTM current is not parked correctly the YIG sphere will interfere with the Band 0 passband.?? This should be fairly straightforward to check by measuring the coil voltage around 4GHz and confirming it parks at a similar location when a Band 0 frequency is selected. -- Wayne
|
Re: Hp 8590a eeprom problem.
I did recap the PS with high grade caps . I'll double check those rails and dig into it and see if I can get a part number for U21.?
|
Re: Hp 8590a eeprom problem.
On 9/7/22 20:42, Richard Cook wrote: I'm getting a error code of 1310 upon turning the SA on. It's indicating eeprom u21. There is some hand writing in the manual (looks like corrections to the manual) that if you get that error code replace corresponding EEPROM. Which is U21.
?Now that code doesn't appear everything I turn it on. Maybe about 75% of the time , and it didn't start giving me that code until I had ran it about 5 hours over 3 days. I got it off ebay and it may have been sitting for a while. .any ideas would be helpful, I have very little knowledge on SA's . It's interesting that it's so intermittent. Have you verified that the instrument's power supplies are stable? What type of EEPROM is U21? (chip part number) Do you know offhand what's stored in it? -Dave -- Dave McGuire, AK4HZ New Kensington, PA
|
Re: Hp 8590a eeprom problem.
I'm getting a error code of 1310 upon turning the SA on. It's indicating eeprom u21. There is some hand writing in the manual (looks like corrections to the manual) that if you get that error code replace corresponding EEPROM. Which is U21.?
?Now that code doesn't appear everything I turn it on. Maybe about 75% of the time , and it didn't start giving me that code until I had ran it about 5 hours over 3 days. I got it off ebay and it may have been sitting for a while. .any ideas would be helpful, I have very little knowledge on SA's .
Thanks
|
Peter ONeill
Jared Cabot
Sergey Kubushyn
Anders
Dave McGuire