|
Hi! In my quest to repair my SA, I am now facing a potential failed input attenuator.?
The dbm readings are within specs as long as I use a multiple of 20 in the attenuator setting. For example, -30,-50 attenuate the signal way more than expected. I read about defective o rings that secure the plugers/pustons to the solenoids. I even watched a video of someone repjacing them.
Would failed o rings cause this failure I am describing?
|
At 08:20 PM 18/01/2022 -0800, you wrote: The dbm readings are within specs as long as I use a multiple of 20 in
the attenuator setting. For example, -30,-50 attenuate the signal way more
than expected. I read about defective o rings that secure the
plugers/pustons to the solenoids. I even watched a video of someone
repjacing them.Would failed o rings cause this failure I am describing? This video? HP RF Attenuators: o-ring repair 'Failed o-rings' means stuck o-rings, that prevent some actuator plungers from moving. Hence wrong attenuations.
|
The symptoms your unit shows point in fact to the O-ring problem. The rubber becomes hard and brittle over the time. I had a couple of attenuators showing the same an fixed them all with 2x1x0.5mm watch crown O-rings. They come from a 'RUIHUA No.4031' set off eBay (yellow-orange sticker on the box). Works perfectly. Just follow the instructions and be careful what you do.
Chris
|
|
|
Awesome. Kit is ordered and I have disassembled the ATT from the main chassis.
|
While disassembly the ATT, I broke the very tip of a plunger. Anyone here have a spare defective ATT then can part with ?
Also, all o-rings look good, so it seems I've broke it for nothing.
Today is not a good day.
|
Nicolas,
While you have it open, examine the attenuator elements under magnification.? You may find that the 10dB element has been damaged by excess power.? If so, you might want to look for a spare attenuator that has a good 10dB element as well as good plungers.
--John Gord
toggle quoted message
Show quoted text
On Wed, Jan 19, 2022 at 02:01 PM, <nicolas.mailloux@...> wrote:
While disassembly the ATT, I broke the very tip of a plunger. Anyone here have a spare defective ATT then can part with ?
Also, all o-rings look good, so it seems I've broke it for nothing.
Today is not a good day.
|
Thank you for the suggestion. I do not see anything burned or blemished.
Please have a look at the pictures :
|
I took some measurements and I think I found something interesting.
Probing resistance from the contact pads to the aluminum block, where the pads are bolted to, I get these readings :
--> Infinite ohm
--> Measure 51.11 ohm
--> Measure 49.9 ohm
--> Measure 61,2 ohm
|
I figured this guy looks like a simple resistor placed in-line and not connected to chassis.
|
Nicolas,
That is a capacitor which can be switched in for AC coupling.? Low frequency response is reduced in AC mode.? (I have seen a cal lab reject an 8594E due to poor low frequency response because they did not realize it was in AC coupled mode.)
Note that the attenuator can be installed reversed.? If that happens, the AC cap is at the wrong end and the input connector does not read as a DC open in AC coupled mode.
--John Gord
toggle quoted message
Show quoted text
On Wed, Jan 19, 2022 at 02:54 PM, <nicolas.mailloux@...> wrote:
I figured this guy looks like a simple resistor placed in-line and not connected to chassis.
|
Do you happen to know the value of said capacitor ? Just for checking it.
Also, do you find anything suspicious in the pictured I posted ?
|
On Wed, Jan 19, 2022 at 03:35 PM, <nicolas.mailloux@...> wrote:
Do you happen to know the value of said capacitor ? Just for checking it.
Also, do you find anything suspicious in the pictured I posted ?
------ According to SM cap is 0.047uF/25V Could be a late suggestion but did you try the attenuator on the bench? You can feed in a 1MHz signal (50-ohm output) and observe the attenuation at the other end ?(50-ohm terminated) without opening the box. You can activate relay contacts one by one on the bench with a power supply and check for steps. A bad relay driver can give similar symptoms. When you observe your signal to be off, is it exactly off by an attenuator step? More than expected or less than expected? Attenuator sections look like pi-attenuators. There are many online calculators that will tell you the expected resistances.? Ozan
|
Ho Ozan,
I have a signal generator, the lower I can go is 14Mhz.
I would like to know where I can find the pinout of the connector. I will manually feed the solenoids.
All I have on hand is my 100Mhz oscilloscope, which should be enough to measure the attenuation by using the Vpp measurement at 14Mhz.
I will of course terminate the connection from the ATT to the DSO with a 50ohm.
I regret not trying it on the bench first before opening it up. Now I do have a broken plunger (trying to fix it with glue...).
|
On Wed, Jan 19, 2022 at 05:30 PM, <nicolas.mailloux@...> wrote:
Ho Ozan,
I have a signal generator, the lower I can go is 14Mhz.
I would like to know where I can find the pinout of the connector. I will manually feed the solenoids.
All I have on hand is my 100Mhz oscilloscope, which should be enough to measure the attenuation by using the Vpp measurement at 14Mhz.
I will of course terminate the connection from the ATT to the DSO with a 50ohm.
I regret not trying it on the bench first before opening it up. Now I do have a broken plunger (trying to fix it with glue...).
------ Higher frequency is fine, the attenuator should work 300kHz-2.9GHz in AC coupling and pretty much down to DC without AC coupling.? 8590A CLIP manual is in the files section, pinout could be the same. I will dig up 8594E attenuator pinout and send in a few hours unless someone else have it handy. Ozan
|
Since you have the long cover off, f you look closely you¡¯ll see the mechanical mechanism which moves the contacts. You should be able to gently push those to toggle between open and closed, as they actually latch mechanically. These are typically pulse driven, not DC. Manually moving them you will see what I mean. Much easier to test and no damage risk to coils or any on-board electronics. ?
toggle quoted message
Show quoted text
From: [email protected] <[email protected]> On Behalf Of nicolas.mailloux@...
Sent: Wednesday, January 19, 2022 8:31 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] 8594E input attenuator issue?? Ho Ozan,
I have a signal generator, the lower I can go is 14Mhz.
I would like to know where I can find the pinout of the connector. I will manually feed the solenoids.
All I have on hand is my 100Mhz oscilloscope, which should be enough to measure the attenuation by using the Vpp measurement at 14Mhz.
I will of course terminate the connection from the ATT to the DSO with a 50ohm.
I regret not trying it on the bench first before opening it up. Now I do have a broken plunger (trying to fix it with glue...).
|
Yes, good idea. However, I still have a broken plunger right now. Doing some 3D printing test...
|
|
|
On Wed, Jan 19, 2022 at 06:19 PM, Ozan wrote:
On Wed, Jan 19, 2022 at 05:30 PM, <nicolas.mailloux@...> wrote:
Ho Ozan,
I have a signal generator, the lower I can go is 14Mhz.
I would like to know where I can find the pinout of the connector. I will manually feed the solenoids.
All I have on hand is my 100Mhz oscilloscope, which should be enough to measure the attenuation by using the Vpp measurement at 14Mhz.
I will of course terminate the connection from the ATT to the DSO with a 50ohm.
I regret not trying it on the bench first before opening it up. Now I do have a broken plunger (trying to fix it with glue...).
------
Higher frequency is fine, the attenuator should work 300kHz-2.9GHz in AC coupling and pretty much down to DC without AC coupling.?
8590A CLIP manual is in the files section, pinout could be the same. I will dig up 8594E attenuator pinout and send in a few hours unless someone else have it handy.
Ozan
-------- I attached a page from the SM and also driver schematic. If you can send picture of the relay section we can tell if there is a contact that disconnects the relay after switching, then it is OK to apply constant DC because current path will be disconnected inside the relay. Some HP relays are built with disconnect switch inside. However, if mechanical actuation is possible it is safest. If it works you will still want to make sure relays operate fine. You can measure coil resistances, then push the plunger(s) and observe if set or reset position of the relevant section goes open circuit confirming built in current disconnect. Let me know if this needs more explanation. Ozan
|
Alright, I tried myself with a 3D printed plunger. It's not perfect, I would need to trim a bit on the top because it keeps the blade too low when engaged, but it's doable with time. The broken plunger is at the capacitor. I booted up my vintage TF2015 Marconi RF generator, set it to 50Mhz and 15Mhz here are the results : PIN 1 = DC BLOCK PIN 2 = 20 dB PIN 3 = 40 dB PIN 4 = 10 dB I fed +/- -20dBm to the ATT from the TF 2015. My DSO read +/-63.2mVpp without the ATT and +/-60mVpp with the ATT all position retracted (OFF), a loss of about 1.5-2mVpp
| Source |
20dB |
40dB |
10dB |
Reading |
| 63.2mVpp |
OFF |
OFF |
OFF |
60mVpp |
| ? |
ON |
OFF |
OFF |
6.40mVpp |
| ? |
OFF |
ON |
OFF |
1.60mVpp |
| ? |
OFF |
OFF |
ON |
19mVpp |
| ? |
ON |
ON |
OFF |
0.800 - 1.2mVpp |
| ? |
ON |
ON |
ON |
0.800 - 1.2mVpp |
| ? |
ON |
OFF |
ON |
2.4-2.8mVpp |
| ? |
OFF |
ON |
ON |
.800-1.2mVpp |
|
Guy Dunphy
