Has and had various issues, mostly work, but WIP to resolve the last remaining one.

⚓ Issues

• FAIL 0308 (resolved)
• Background RF noise (wip)

⚓ Troubleshoot

⚓ Initial

From the Auction picture

  HP-IB ADRS: 18
Copyright HP 1986
8590A rev 14.10.88
FAIL: 0308 0000000000

Fail code:
0308 code according to one documentation is:

0 N/A
3 A16 led 9; Ram Odd FF4000; On during normal operation
  A16 led 8; Ram Even FF4000; On during normal operation
0 N/A
8 A16 led 3; ROM ODD B LSB; IC U6

The Fail code then indicates an issue with ROM Checksum. The RAM Code indicates RAM is present at FF4000.

EPROM are U6, U7, U23, U24. PROM 32K X8. 256K.

Also have noise on the RF.

⚓ 15/06/2020

Error code 0308 appears after reset. Calibration passes, then after sometimes the error code 1308 appears instead.

1308:

68230
ram odd FF4000 (expected)
ram even FF4000 (expected)
N/a
rom odd b lsb U6

Eeproms are 1818-3329, 256K and has label with Rev F from 1989.

⚓ 16/06/2020

Got firmware dumps from http://bluefeathertech.com/

My U6 EPROM was checksumming at 5C00, the right one is 5C01, one naughty byte got eaten by a wild cosmic ray.

Flashed a new 27C256 with right dump, no more error !

The RF is still wacked up but it’s a (small) progress.

⚓ 11/07/2020

First converter has “-23dBm” handwritten on it, mmh.

Diodes of it were replaced by a single HSMS2822.

Calibration data is the absolute same, nothing better, before and after:

One thing to note, cal off, signal is clean, cal on, the signal has a weird dip in the middle of it.

Cal on vs Cal off:

Signal generator output -65dBm, spectrum analyzer displays -53dBm, ±-12dBm seems ≈consistent between 10MHz and 1GHz.

Spectrum analyzer is off by ±2MHz.

⚓ 11/01/2021

Did some measurements.

Cal output should be OK at ~300MHz.

2ND LO is at 1730.6MHz (service manual says 1728.7MHz+-0.5MHz, so let’s say it’s ok)

Calibration still not much better.

To check: https://web.archive.org/web/20201127233323/https://community.keysight.com/thread/5422

⚓ 04/08/2021

More tests, checks done. Everything looks ok, powers and frequencies are good.

• https://www.eevblog.com/forum/repair/hp-8590a-wrong-frequency-and-amplitude-measurements-below-17mhz/
• https://www.eevblog.com/forum/beginners/first-spectrum-analyzr-hp-8590a/

Did one more test.

Injected 200MHz at -20dBM into a tinySA, reading of -20.4dBm, switched that signal into RF IN of the first mixer, IF OUT to the tinySA, reading of -71dBm, so the first mixer is eating ~50dBm.

⚓ 12/04/2022

Received a second hand unknown state First Converter and Second Converter.

First Converter has been tested, -20dBM in RF IN, around ~-71dBm onIF OUT, so if this FC is ok, I get the same measurements as my repaired one.

Second hand First and Second converters don’t change anything.

Still having the same garbled datas. However I have fed 299.9MHz from a known calibrated generator, and calibrated the SA 2ND Lo Frequency (C4) so there is no more a +-2MHz offset. Also re-calibrated the calibrator output after doing that.

Running [CAL][MORE][CONF TEST] renders a… lot of errors, at least:

POS-PK FAIL
SAMPLE FAIL
POS-PK FAIL
VID-BW FAIL    100000
VID-BW FAIL     30000
VID-BW FAIL     10000
VID-BW FAIL      1000
RES-BW NOISE FAIL   3000000
RES-BW NOISE FAIL   1000000
RES-BW NOISE FAIL    300000
RES-BW NOISE FAIL    100000
RES-BW NOISE FAIL     30000
RES-BW NOISE FAIL     10000
RES-BW NOISE FAIL       3
RES-BW NOISE FAIL       1000
STEP GAIN/ATTN FAIL  10
STEP GAIN/ATTN FAIL  20
RES-BW SHAPE FAIL   3000000
RES-BW SHAPE FAIL   1000000
RES-BW SHAPE FAIL    300000
RES-BW SHAPE FAIL    100000
RES-BW SHAPE FAIL     30000
RES-BW SHAPE FAIL     10000