¿ªÔÆÌåÓý

HP 8664A signal generator repairability


 

Hi everyone,

I recently purchased an 8664A signal generator with option 004 and 10. Got a decent deal. Unfortunately, it appears that UPS dropped the box and despite plenty of foam packing, the instrument sustained some pretty severe damage.

  1. Functionally, the instrument seems to be performing quite well. I tested it on my 8566A spectrum analyzer. No issues here.
  2. The right hand rack handle appears to have taken the brunt of the damage. It's destroyed. The left handle is slightly bent, but not broken.
  3. Unfortunately, it did not save the RF output N connector. This is badly bent. It's still connected inside as I'm getting output.

The seller is willing to work with me if I want to send it back for a full refund, but in the meantime, I want to try to estimate what repair work might cost me. I can get another set of 4u rack handles for $40 or so on ebay, so that's not a real problem. My main concern is the RF output and the front panel. I'd like to get into the instrument to ascertain the damage, but I can't seem to figure out how open the case. The service manual provided by Keysight talks about removing top and bottom panels, but mine seems to have a 1 piece case that doesn't want to just slide off.

To conclude, my major questions are:

  1. Is it worth fixing the damage?
  2. How do you open the case?

Thanks!

Sean


 

Also, on initial power up, it did the lengthy self cal routine (dead battery?) and when it finished it flashed up code 8043311 and had a message "hardware failure 8". I think this is related to frequency ref? Should I have plugged in an external reference? The error message disappeared and hasn't shown up again.


 

On 3/18/19 7:03 PM, [email protected] wrote:
I recently purchased an 8664A signal generator with option 004 and 10.
Got a decent deal. Unfortunately, it appears that UPS dropped the box
and despite plenty of foam packing, the instrument sustained some pretty
severe damage.

1. Functionally, the instrument seems to be performing quite well. I
tested it on my 8566A spectrum analyzer. No issues here.
2. The right hand rack handle appears to have taken the brunt of the
damage. It's destroyed. The left handle is slightly bent, but not
broken.
3. Unfortunately, it did not save the RF output N connector. This is
badly bent. It's still connected inside as I'm getting output.


The seller is willing to work with me if I want to send it back for a
full refund, but in the meantime, I want to try to estimate what repair
work might cost me. I can get another set of 4u rack handles for $40 or
so on ebay, so that's not a real problem. My main concern is the RF
output and the front panel. I'd like to get into the instrument to
ascertain the damage, but I can't seem to figure out how open the case.
The service manual provided by Keysight talks about removing top and
bottom panels, but mine seems to have a 1 piece case that doesn't want
to just slide off.

To conclude, my major questions are:

1. Is it worth fixing the damage?
I would say so, yes. Those are, in my opinion, some of the finest
oscillators HP ever built. Handles are no big deal, if you actually
even want them on the machine. I'd replace, rather than try to repair,
the RF connect.

2. How do you open the case?
Remove the rear feet. Next, remove the long handles from the sides.
This will expose a row of screws running from front to back on each
side. They are usually Torx. Remove all of those, and the top should
come off by sliding it back slightly and then lifting up.

...and after typing that, I just re-read what you said above about it
being a one-piece case. The long handles on the sides cover the row of
screws that I described above; are you certain it's one piece?

-Dave

--
Dave McGuire, AK4HZ
New Kensington, PA


 

Dave,

Thanks for getting back. There are no long handles on the case. I am familiar with them because all my other HP gear has them. This case is different. See attached.


 

On Mon, Mar 18, 2019 at 7:03 PM <[email protected]> wrote:
  1. Unfortunately, it did not save the RF output N connector. This is badly bent. It's still connected inside as I'm getting output.

I recently fixed a HP8665A which had taken a faceplant, such that the output connector was MIA. I bought a replacement part from Keysight, but their "upgraded part" didn't match the original and didn't fit. I ended up buying a couple of these [] NOS replacement connectors off eBay (for less money than the supposed-original Keysight part).
I had to machine a new mounting bracket as mine was MIA. I used the CNC mill, but if the mounting bracket is too far gone, you can make a new one from angle iron (or alu) with a drill press and a file.

I worried about the output semi-rigid coax, which had taken some of the hit. The solder connection to the SMA connector had cracked, and there was a bit of a bend in the coax. I bent the coax back and re-soldered the shield and if there's any problem with the output line I can't measure it.

The battery message and self-cal are normal. The battery is a 3.6V NiCad that can only power the memory for a couple of weeks or so at best. I replaced the battery as it was only holding the memory up for a day or so, and I understand they can leak, which in turn can damage the motherboard.


 

On Mon, Mar 18, 2019 at 04:35 PM, Siggi wrote:
On Mon, Mar 18, 2019 at 7:03 PM <[email protected]> wrote:
  1. Unfortunately, it did not save the RF output N connector. This is badly bent. It's still connected inside as I'm getting output.
?
I recently fixed a HP8665A which had taken a faceplant, such that the output connector was MIA. I bought a replacement part from Keysight, but their "upgraded part" didn't match the original and didn't fit. I ended up buying a couple of these [] NOS replacement connectors off eBay (for less money than the supposed-original Keysight part).
I had to machine a new mounting bracket as mine was MIA. I used the CNC mill, but if the mounting bracket is too far gone, you can make a new one from angle iron (or alu) with a drill press and a file.
?
I worried about the output semi-rigid coax, which had taken some of the hit. The solder connection to the SMA connector had cracked, and there was a bit of a bend in the coax. I bent the coax back and re-soldered the shield and if there's any problem with the output line I can't measure it.
?
The battery message and self-cal are normal. The battery is a 3.6V NiCad that can only power the memory for a couple of weeks or so at best. I replaced the battery as it was only holding the memory up for a day or so, and I understand they can leak, which in turn can damage the motherboard.
As the pictures above show, I was able to at least get the front panel off. It appears that the mount you speak of took the brunt...the N connector and the hardline behind it look fine. I bent it back with the help of some scrap leather and heavy duty lineman pliers.

Thanks for the battery info!

Sean


 

Peel the decorative vinyl strips off of the collar just behind the front of the case.? Beneath them, you will find a gazillion screws.? (ok, that's a slight exaggeration - there are 48 of them.)? Remove said screws, along with the bumper feet on the back of the instrument (if present), and the sleeve case will then slide off.


On Mon, Mar 18, 2019 at 07:34 PM, [email protected] wrote:

Dave,

Thanks for getting back. There are no long handles on the case. I am familiar with them because all my other HP gear has them. This case is different. See attached.


 

On Mon, Mar 18, 2019 at 04:44 PM, Patrick Manning wrote:
Peel the decorative vinyl strips off of the collar just behind the front of the case.? Beneath them, you will find a gazillion screws.? (ok, that's a slight exaggeration - there are 48 of them.)? Remove said screws, along with the bumper feet on the back of the instrument (if present), and the sleeve case will then slide off.
OH, I see them now. Thanks!! This is going to be frustrating! The strips are peeling up already anyway, so good riddance.


 

IIRC I went and got my impact driver to take them out after doing a few by hand.? Mine had threadlocker on them for extra fun; I'd expect yours will, too.? (I put them back in by hand once the instrument was fixed.)

-Pat


On Mon, Mar 18, 2019 at 07:53 PM, [email protected] wrote:

On Mon, Mar 18, 2019 at 04:44 PM, Patrick Manning wrote:
Peel the decorative vinyl strips off of the collar just behind the front of the case.? Beneath them, you will find a gazillion screws.? (ok, that's a slight exaggeration - there are 48 of them.)? Remove said screws, along with the bumper feet on the back of the instrument (if present), and the sleeve case will then slide off.
OH, I see them now. Thanks!! This is going to be frustrating! The strips are peeling up already anyway, so good riddance.


 

On 3/18/19 7:34 PM, [email protected] wrote:
Thanks for getting back. There are no long handles on the case. I am
familiar with them because all my other HP gear has them. This case is
different. See attached.
Wow, bizarre...I've never seen a case like that on an 866x. I see
someone else has gotten you fixed up. You might not be in such a bad
situation after all; maybe put the screws to the seller about the
shipping and get half of your money back. :)

-Dave

--
Dave McGuire, AK4HZ
New Kensington, PA


 

That case is Option 10...low leakage. I have one.... It is rare, indeed


On Mon, Mar 18, 2019, 4:58 PM Dave McGuire <mcguire@...> wrote:
On 3/18/19 7:34 PM, [email protected] wrote:
> Thanks for getting back. There are no long handles on the case. I am
> familiar with them because all my other HP gear has them. This case is
> different. See attached.

? Wow, bizarre...I've never seen a case like that on an 866x.? I see
someone else has gotten you fixed up.? You might not be in such a bad
situation after all; maybe put the screws to the seller about the
shipping and get half of your money back. :)

? ? ? ? ? ? ?-Dave

--
Dave McGuire, AK4HZ
New Kensington, PA




 

8644 had the same case for some options, I believe this was on the
later instruments? All I have seen says Agilent, instead of HP.

FWIW, Some instruments in that series was buildt by the now quite
famous "Screw murderer", and as such had screws threaded down at a
angle, stripping the threads and causing all kinds of havock when you
are trying to get them out, if they don't break in the hole.

BR.
Thomas.

Den tir. 19. mar. 2019 kl. 00:58 skrev Dave McGuire <mcguire@...>:


On 3/18/19 7:34 PM, [email protected] wrote:
Thanks for getting back. There are no long handles on the case. I am
familiar with them because all my other HP gear has them. This case is
different. See attached.
Wow, bizarre...I've never seen a case like that on an 866x. I see
someone else has gotten you fixed up. You might not be in such a bad
situation after all; maybe put the screws to the seller about the
shipping and get half of your money back. :)

-Dave

--
Dave McGuire, AK4HZ
New Kensington, PA


--
With Best regards, Thomas S. Knutsen.

Please avoid sending me Word or PowerPoint attachments.


 

Dave, Option 10 is a 'reduced leakage' configuration that does away with the normal top and bottom panels and side handles we all know and love (certainly from the removal standpoint!) and replaces them with a solid sleeve case with said gazillion screws to reduce RF leakage.


Pics of the 8644B I have are at:





-Pat

On Mon, Mar 18, 2019 at 07:58 PM, Dave McGuire wrote:

On 3/18/19 7:34 PM, [email protected] wrote:
Thanks for getting back. There are no long handles on the case. I am
familiar with them because all my other HP gear has them. This case is
different. See attached.
Wow, bizarre...I've never seen a case like that on an 866x. I see
someone else has gotten you fixed up. You might not be in such a bad
situation after all; maybe put the screws to the seller about the
shipping and get half of your money back. :)

-Dave

--
Dave McGuire, AK4HZ
New Kensington, PA



 

On Mon, Mar 18, 2019 at 05:03 PM, Thomas S. Knutsen wrote:
8644 had the same case for some options, I believe this was on the
later instruments? All I have seen says Agilent, instead of HP.

FWIW, Some instruments in that series was buildt by the now quite
famous "Screw murderer", and as such had screws threaded down at a
angle, stripping the threads and causing all kinds of havock when you
are trying to get them out, if they don't break in the hole.

BR.
Thomas.
Well, so far the front panel screws were all ok. I'll keep that in mind. Thanks!


 

¿ªÔÆÌåÓý

Sean,

?

Exactly the same thing happened to me with a HP8664A - no handles but with options in mint condition with a stated ¡°power supply¡± fault.

It was packed badly and sustained bad damage to the front lower left hand side front panel and had the two bottom rear base rubber stops damaged or broken off.

The Impact to the front bent the front edge of the outer case meaning I had to use force to get the case off.? I cut away the bent casting to straighten the face plate.

?

My familiarity with the same issue to failing tantalum capacitors in an HP8665A helped me sort that the power supply fault out.

and also replaced the backup battery.? Unfortunately at power on, the unit only seemed to frequency lock over a very small range.

From then on, the unit was a nightmare with errors from turn on until I realised they had left the internal switch in service mode and got nowhere because the display was not active.

All good now.? The internal oscillator takes time to warm up from when AC power is applied.? Using an external reference makes it good to go much quicker.

?

Fixing ?yours?

?

Your generator sounds like is it pretty good if you can fix the cosmetics. ?The handles protected the front keypads and the RF connector/ assembly can be replaced.

?

They are well worth the repair efforts.

Keep track of where you take out the screws from.? They use variations of length and tapering and there are lots of them.

?

Regards Noel

From: [email protected] [mailto:[email protected]] On Behalf Of [email protected]
Sent: Tuesday, 19 March 2019 10:03 AM
To: [email protected]
Subject: [HP-Agilent-Keysight-equipment] HP 8664A signal generator repairability

?

Hi everyone,

I recently purchased an 8664A signal generator with option 004 and 10. Got a decent deal. Unfortunately, it appears that UPS dropped the box and despite plenty of foam packing, the instrument sustained some pretty severe damage.

  1. Functionally, the instrument seems to be performing quite well. I tested it on my 8566A spectrum analyzer. No issues here.
  2. The right hand rack handle appears to have taken the brunt of the damage. It's destroyed. The left handle is slightly bent, but not broken.
  3. Unfortunately, it did not save the RF output N connector. This is badly bent. It's still connected inside as I'm getting output.


The seller is willing to work with me if I want to send it back for a full refund, but in the meantime, I want to try to estimate what repair work might cost me. I can get another set of 4u rack handles for $40 or so on ebay, so that's not a real problem. My main concern is the RF output and the front panel. I'd like to get into the instrument to ascertain the damage, but I can't seem to figure out how open the case. The service manual provided by Keysight talks about removing top and bottom panels, but mine seems to have a 1 piece case that doesn't want to just slide off.

To conclude, my major questions are:

  1. Is it worth fixing the damage?
  2. How do you open the case?


Thanks!

Sean


Virus-free.


 

On Mon, Mar 18, 2019 at 04:08 PM, Dave McGuire wrote:
On 3/18/19 7:03 PM, [email protected] wrote:
I recently purchased an 8664A signal generator with option 004 and 10.
Got a decent deal. Unfortunately, it appears that UPS dropped the box
and despite plenty of foam packing, the instrument sustained some pretty
severe damage.

1. Functionally, the instrument seems to be performing quite well. I
tested it on my 8566A spectrum analyzer. No issues here.
2. The right hand rack handle appears to have taken the brunt of the
damage. It's destroyed. The left handle is slightly bent, but not
broken.
3. Unfortunately, it did not save the RF output N connector. This is
badly bent. It's still connected inside as I'm getting output.


The seller is willing to work with me if I want to send it back for a
full refund, but in the meantime, I want to try to estimate what repair
work might cost me. I can get another set of 4u rack handles for $40 or
so on ebay, so that's not a real problem. My main concern is the RF
output and the front panel. I'd like to get into the instrument to
ascertain the damage, but I can't seem to figure out how open the case.
The service manual provided by Keysight talks about removing top and
bottom panels, but mine seems to have a 1 piece case that doesn't want
to just slide off.

To conclude, my major questions are:

1. Is it worth fixing the damage?
I would say so, yes. Those are, in my opinion, some of the finest
oscillators HP ever built. Handles are no big deal, if you actually
even want them on the machine. I'd replace, rather than try to repair,
the RF connect.
Thanks! After opening the front panel, I discovered that the mount was simply bent. I bent it back with some scrap leather for protection and linesman pliers. The hardline coax behind appears to be fine. Not seeing any noise coming in through to the spectrum analyzer so far.

I think, perhaps, I may escape from this simply by purchasing a new to me set of 4U rack handles.

Sean


 

On Mon, Mar 18, 2019 at 05:46 PM, Noel VK3NH wrote:

Sean,

?

Exactly the same thing happened to me with a HP8664A - no handles but with options in mint condition with a stated ¡°power supply¡± fault.

It was packed badly and sustained bad damage to the front lower left hand side front panel and had the two bottom rear base rubber stops damaged or broken off.

The Impact to the front bent the front edge of the outer case meaning I had to use force to get the case off.? I cut away the bent casting to straighten the face plate.

?

My familiarity with the same issue to failing tantalum capacitors in an HP8665A helped me sort that the power supply fault out.

and also replaced the backup battery.? Unfortunately at power on, the unit only seemed to frequency lock over a very small range.

From then on, the unit was a nightmare with errors from turn on until I realised they had left the internal switch in service mode and got nowhere because the display was not active.

All good now.? The internal oscillator takes time to warm up from when AC power is applied.? Using an external reference makes it good to go much quicker.

?

Fixing ?yours?

?

Your generator sounds like is it pretty good if you can fix the cosmetics. ?The handles protected the front keypads and the RF connector/ assembly can be replaced.

?

They are well worth the repair efforts.

Keep track of where you take out the screws from.? They use variations of length and tapering and there are lots of them.

?

Regards Noel

WOW!! That corner got violated. I bet if mine didn't have the rack handles installed, it would have ended up exactly that way. Thanks for the warning about the screws.

Sean


 

make sure you check the output both below and above 375MHz and at different levels, too. to make sure the mechanical attenuator is OK
?
i think error 8 is not unusual if the OVEN is still cold and/or the OCXO is not stabilized yet
even if it is TCXO still the same. If the error goes away, that's not a big deal.


On Mon, Mar 18, 2019 at 07:07 PM, @0culus wrote:
Also, on initial power up, it did the lengthy self cal routine (dead battery?) and when it finished it flashed up code 8043311 and had a message "hardware failure 8". I think this is related to frequency ref? Should I have plugged in an external reference? The error message disappeared and hasn't shown up again.


 

Very interesting, I didn't know about this! Thanks for the info.

-Dave

On 3/18/19 8:01 PM, Colby Burkett wrote:
That case is Option 10...low leakage. I have one.... It is rare, indeed

On Mon, Mar 18, 2019, 4:58 PM Dave McGuire <mcguire@...
<mailto:mcguire@...>> wrote:

On 3/18/19 7:34 PM, [email protected] <mailto:[email protected]> wrote:
> Thanks for getting back. There are no long handles on the case. I am
> familiar with them because all my other HP gear has them. This case is
> different. See attached.

? Wow, bizarre...I've never seen a case like that on an 866x.? I see
someone else has gotten you fixed up.? You might not be in such a bad
situation after all; maybe put the screws to the seller about the
shipping and get half of your money back. :)

? ? ? ? ? ? ?-Dave

--
Dave McGuire, AK4HZ
New Kensington, PA



--
Dave McGuire, AK4HZ
New Kensington, PA


 

On 3/18/19 8:08 PM, Patrick Manning wrote:
Dave, Option 10 is a 'reduced leakage' configuration that does away with
the normal top and bottom panels and side handles we all know and love
(certainly from the removal standpoint!) and replaces them with a solid
sleeve case with said gazillion screws to reduce RF leakage.


Pics of the 8644B I have are at:


I'm drooling here. :) I have an 8644A. The *only* thing I don't like
about it is that to see what amplitude it's configured for, I have to
turn the RF on. Other than that...I just adore the thing. Great
interface, great functionality, great output..

-Dave

--
Dave McGuire, AK4HZ
New Kensington, PA