EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: Calvin on December 30, 2019, 06:24:04 am
-
Hi,
I´ve the chance to get a HP3577A in an auction.
The initial bootup runs all ok. But the confidence tests list quite a number of errors on all three inputs.
As I understand all the same messages on all thee inputs and two more on Input R.
Unfortunately one can´t read the error messages, as the pics are so small.
Anyway, do those confidence test messages mean that the HP is just out of tune/calibration or are there also messages possible indicating major issues?
What would be a resonable price for a unit that looks visibly well, without scratches and low sign of wear and a nice display, that boots up wo issues ... >300€ or less?
Thanks alot
Calvin
-
Hi Calvin
Thanks for your message.
The messages during the self confidence test seem to be the same, only the passed / failed identifier changes.
Attached is a picture of mine after the tests for you to compare...
I haven't looked in the manual, but since your unit fails receiver impedance test I'm worried. That sounds like a hardware fault to me...
I paid about 800USD with 50Ohm S-Parameter test set with the unit passing all tests and sold as fully working. Plus about 600USD worth of shipping and taxes :palm:
-
Hi,
thanks for the screenshot :-+
Upsizing my pics and comparing with Your screenshot it can indeed be identified as ´Receiver impedance test´ fail on all three inputs and a ´Log´ and a ´Linear sweep signal level test´ on input R.
The last sound like they are decalibration issues.
Thank You so much so far
Calvin
btw. great new display on Your machine :-+
-
btw. great new display on Your machine :-+
Yes, day and night to the phosphor screen. Pricy, but the digital interface which hp used to drive the display a diy solution would be a huge pain.
Altought often I get the data via GPIB and do a lot of processing with python scripts so you might get around the screen entirely... Depends on how you want to use it!
-
The level test error on the R channel is probably just an adjustment; the sweep shape is approximately correct.
The impedance error could be several things and needs to be looked into. The most common fault I've seen on the receivers is the 50 Ohm input attenuator resistors being blown, they're only 1/8 watt, and are engaged by default. Max input level on these things is pretty low. If the channel levels are proper in 1M mode but fail in 50 Ohm, that would be the first thing to look at.
300 Euro for a unit that is known to need work sounds very high. If the crt is strong and sharp, I'd go 1/2 to 1/3 that much.
-
Just a guess but I remember that the 3577 has a bunch of RF Relays in the front ends. They are mechanical and if they wear out they might change impedance. If so your only option is to buy another device used which has them since they are around 100$ a piece new.
-
Hi,
today I a took the chance to check my 3577A.
Whoaw what a build! It felt like almost 2kg of screws sticking in the top of the device alone ... multiple screened chambers from thick aluminum ... well it´s obvious now where the weight came from.
Seemed that it hadn´t been opened once in all the years before.
Anyway, after studying the (downloaded) manual found out that the most important parts were missing ....the pages with the input boards schematic and board stuffung |O
So I used the (downloaded) 3577B manual and checked the resistors defining the input impedances and found them ok for 1M and 50R on all three boards.
Checked the manual for a signal level test procedure and learned that all three inputs were off, between 1M-level and 50R-level.
The 3577A checks if both levels differ within 6dB +-0.2dB.
Secondly it checks the level difference with or without 20dB attenuation (less than +-0.1dB).
Channel R was off at 1M/50R by roundabout 3dB, while Channel A and B were just marginally above the +-0.2dB limit.
Since channel R was so far off it also failed on the Log and Linear sweep tests.
Levels can simply be adjusted by two potis through the top lids of the boards.
But before I checked the signal level coming from the Source channel (-30dBm) against my function generator.
Then fed the source signal into the input channels and adjusted the levels.
Restarted the VNA and voila .... all confidence tests passed .. and I have a working 3577A for 250€. :-+
Will need to clean the beast thoroughly and will probabely have to adjust/trim one or the other parameter that has drifted over the years.
Now I can start the hunt for a reasonable priced 35676A s-Parameter test set. ;)
Many thanks to you helpers :-+
regards
Calvin