Replacing the fan on an HP11729C
I recently needed to change the fan on an HP11729C. Several members of the HP-Agilent-Keysight-equipment list on groups.io greatly assistment me in this, which is in fact a bit of an understatement. Without their help I would not have been able to complete this task. I would like to acknowledge their assistance and thank them for it. They are: Leo Bodnar, Chuck Harris, Orin Eman and John Gord. In the spirit of their unselfish contributions, I thought I would create a detailed set of instructions that document my experience with this retrofit.
The original equipment on the HP11729C is a Buehler fan driven by control logic that converts 12V DC to a 3-phase AC driving voltage. The new fan is a 12V DC fan. The one I used was recommended by Leo - , which I purchased from Amazon.
In order to remove the fan, several bolt sets must be removed. My 11729C has handles on the side for lifting, which must be removed, as their edges cover the backplate in which the fan is mounted (see handles.jpg, which is attached to this message). There are two of these handles, one on the top, the other on the bottom. Then you must remove 8 mini-bolts, 4 on each side of the chasis (see four_backplate_side_bolts.jpg). Note: two of these bolts are longer than the others. You should mark the holes from which the came so they can be returned to the proper place when reattaching the backplate to the chasis.
The final set of bolts are located on the face of the backplage (see backplate_facing_bolts.jpg). These are easily removed, but (see below) are incredibly hard to get back into place. If you try to remove the old fan without removing these bolts, you run the risk of bending the backplate, so unfortunately, their remove is necessary.
Once these set of bolts are removed, with a bit of wiggling the old fan can be wedged out of its slot. This is not particularly easy as the old fan sits in a slot on the side of the backplate (see lip_on_backplate_holding_fan.jpg. Note: this lip is shown in the image with the new fan inserted, so it isn't apparent how the old fan inserts into it. However, when you push the backplate out to access the old fan, it is obvious). After removing the bolts holding the old fan and fan guard to the backplate, you can wiggle it in order to wedge it out of the other side of the backplate (see new_fan_insterted_into_backplate.jpg. Note this image shows the new fan in the fan slot with the fan guard and fan bolted to the backplate. However, the arrows show where the fan can be wedged out of its slot.)
Once you have removed the old fan, the new fan slips easily into place. Just bolt the fan guard and fan to the backplate using the bolts that held the old fan in place.
In order to connect the new fan to its power source, the red and black wires must be routed to a 12V source. This is on the other side of the PC board, so (after twisting the wires) I routed it through a convenient VIA (see site_of_12V_supply_on_back_of_PC_board.jpg). Note that the schematic (see Schematic.jpg) shows the speed control for fan connected to the -15V supply on the 11729C. I orignally did not understand how this was going to provide the necessary 12 volts required by the new fan. However, Chuck Harris pointed out that the PI filter inserted between the -15V supply voltage and the speed control would drop the voltage to around 12V, since it comprises two 19.6 Ohm resistors with a 10 uF cap to ground. In fact when loaded by the new fan, the PI filter drops the voltage to 11.23V. This is sufficient, since the new fan will spin on volages between 6 and 13.8 V.
It is important to note that the new fan runs on 12 V, whereas the speed control expects a -12 V input. So, when wiring the new fan to the old power source, you must solder the red wire to ground and the black wire to -12 V (see 12v_supply_to_fan.jpg). Note that the red wire is soldered to the 2 pad and the black wire to the 6 pad, which, if you look at the schematic accomplishes the required switch in polariy.
Once the new fan is inserted (and after testing that it works by connecting the power cord and turning on the unit), it is time to reattach the backplate to the chasis. I first rebolted the 8 side bolts, connecting the backplate to the chasis side wall. Then I rebolted the facing bolts through the backplate to the chasis. This is an easy statement to make, but it wasn't easy to accomplish.
It took me about 2 1/2 hours to get these two bolts connected. Working on each bolt one at a time, I inserted them with the unit sitting in its side. I then put a piece of electral tape over the bolt hole (see tape_holding_backplate_facing_bolts.jpg) and turned the unit so the bolt face was sitting on the benchtop (see backplate_facing_bolt_on_inside_of_chasis.jpg). This allowed me, using a pair of needle nose pliers, to gently ease the washer and lock washer over the bolt. I then turned the unit on its side again and very gently removed the tape, using my finger to keep the bolt from coming out with the tape. Splitting a wooden tongue depressor lengthwise and adding some blue stickum to its end (see backplate_facing_bolt_nut_on_half_tongue_depressor.jpg) I put the nut on the stickum. Not shown in the image is I then bent the end of the tongue depressor 90 degrees so I could place the nut over the bolt end. Using a screw driver, I turned the bolt wiggling the nut around until it caught on the bolt's threads. I then removed the tongue depressor and using a pair of needle nose pliers, held the nut in place using the screw driver to tighten the nut on the bolt.
This may sound straight forward, but it was not. First, getting the washer and lockwasher on the bolt end was not easy. They kept flying all over the place when I tried to manuever them onto the bolt end with the needle nose pliers.
Second, when I had them in place and turned the unit back on its side, the washers would invariably fall off, requiring me place it on its back again and reposition the washers. I kept loosing them as they fell into the insides of the unit. Turing the unit this way and that, simultaneously shaking it would usually disloged them and either they would fall out onto the bench or would land somewhere in the chasis that I could access. However, I lost one washer that never did come out of the unit's bowels.
Third, when I had successfully got the unit on its side with both washers in place and tried to maneuver the bolt over the bolt end, the stickum would catch the washers and further movement of the tongue depressor would pull them off.
I won't go into any more detail, but be forwarned. Getting these bolts back into place is a very frustrating task. Perhaps others have some advice how to do this with more ease.
Once the facing bolts were reattached, it only requires attaching the side handles and the repair is complete.
Dan
|
Here's another interesting short obit:
On Thursday, February 20, 2020, 3:28:23 PM PST, Tim hughes <timhughes@...> wrote:
He was still very active up to the end.
If you are registered on ResearchGate or just go and google ResearchGate, you can see he was still a very active contributor to discussions on electronics always ready to help and encourage new engineers asking naive questions. You can download some of his papers/presentations from researchgate too.
Apart from all his contributions to Multipliers/mixers and non-linear ccts, I always liked his unique analog circuits for putting the display characters on tektronix analog scopes.
Tim
Very sad news indeed. Thanks for sharing.
Sean
|
He was still very active up to the end.
If you are registered on ResearchGate or just go and google ResearchGate, you can see he was still a very active contributor to discussions on electronics always ready to help and encourage new engineers asking naive questions. You can download some of his papers/presentations from researchgate too.
Apart from all his contributions to Multipliers/mixers and non-linear ccts, I always liked his unique analog circuits for putting the display characters on tektronix analog scopes.
Tim
Very sad news indeed. Thanks for sharing.
Sean
|
Re: sick hp 5421A Digital Processor
Hi Mark,
as what per the functional specification can be assumed, the hp 5480A averages repetitive signals in the amplitude domain only, whereas the 5421A portion of the 5401A is capable of doing much more than that (MCS, PHA, Correlation between two signals etc.). One thing in common is that digital amplitude values are stored in a magnetic core memory. I have no clue if these instruments actually came from the same R&D team (maybe someone else knows), but I assume that nobody at hp wanted to invent the wheel twice. It is not beyond possibility that there are shared components or even assemblies. The hpJ March and April 1968 issues suggest though that both instruments were cranked out by the Nuclear Instrumentation Group.
Do you have the manual in electronic form or as paper hardware? I would be interested in whatever is available.
In return: I have paper versions of the following SMs: 5431A/B, 5416A/B and 5406A (system SM only, just external wiring and troubleshooting diagrams, no schematics) from which I can make (partial) scans if this is needed for another repair project out there. Just let me know!
Chris
|
From time to time I come across some very interesting people who have interests in a similar vein to those of mine.? Recently I was reminded of one particular person ¨C Kenneth A. Kuhn who has a very deep interest in older HP equipment ().? I had come across his website some time in the past and again had been reminded of it by a colleague who had just come upon it.
?
While looking through his web pages I noted that they had not been updated in a few years and became curious as to Mr. Kuhn¡¯s status.? Does anyone know or might he be a member of this mail list?
?
Thanks,
Greg
|
Re: sick hp 5421A Digital Processor
Hi Eagle-Eye Dave? ;-) ,
did not realize that the NIMs were on the picture too. Actually, they were kind of a by-catch when I bought a crate full of Gov Surplus stuff (including the hp 5401A) from one of the the NV atomic testing sites many years ago when I was living part-time time in SoCal. Problem: there was no NIM bin in this lot. I would have liked to get one of these nice hp 5580A NIM bins, but NIM bins are scarce in general here in Germany and finding a hp bin seemed to be impossible. Last year I started to visit the HAM Radio Friedrichshafen here in Germany. Maybe I will find an 'ordinary' Canberra or Ortec bin there some day...
Chris
|
On February 20, 2020 8:40:59 AM "Dave_G0WBX via Groups.Io" <g8kbvdave@...> wrote: Someone did write an open-source Linux program for displaying video from
the Seek cameras. I've used it; it works.
------------------------------------------------------------------------
Go on then, don't just dangle the carrot.? What's the name of that
application, or a link to the authors site or GitHub page.
If it could build & run on a Pi (Other SBC's are available) you could
almost make a dedicated device. Sorry, I figured people would Google it. ;) I'd forgotten where it was, so I Google it. Here you go: I've not tried it on a Pi or similar SBC. Please let me/us know if you do. -Dave -- Dave McGuire, AK4HZ New Kensington, PA
|
From: Dave
McGuire
Date: Wed, 19 Feb 2020 10:26:50 PST
<Snip>
Someone did write an open-source Linux program for displaying
video from the Seek cameras. I've used it; it works.
Go on then, don't just dangle the carrot.? What's the name of
that application, or a link to the authors site or GitHub page.
If it could build & run on a Pi (Other SBC's are available)
you could almost make a dedicated device.
Cheers.
Dave G0WBX.
--
Created on and sent from a Unix like PC running and using free and open source software:
|
set 8350b to 15 mhz cw and did a single sweep on the spectrum analyzer and it produced a similar waveform to the previous picture, just a bit more squashed up.
|
OK- hadn¡¯t spotted that.? But if you can single shot the spec an sweep now, you should see the sweeper signal at various places across the screen each time the spec an sweeps. Should be ?a single signal each time. ( assuming all ¡®memorizing¡¯ functions are turned off-eg peak hold, etc)? If you still see multiple signals the sweeper probably does have a problem. DaveB, NZ ? ?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of Ken Goodhew
Sent: Thursday, February 20, 2020 21:25
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] hp 83592A problem? Hi Dave, ??????????????? The sweep rate for the sweeper was 10 seconds for 10 mhz to 1.8ghz as depicted in the picture ? ? What do you see when you slow the 8350 sweep rate down to be much slower than the sweep rate of the spec an? DaveB, NZ ? ? From: [email protected] [mailto:[email protected]] On Behalf Of Ken Goodhew
Sent: Thursday, February 20, 2020 18:04
To: [email protected]
Subject: [HP-Agilent-Keysight-equipment] hp 83592A problem ? Hi Gang,
I have an 8350b fitted with an 83592A plug in.
The problem exhibits as no control over the output power,but the unleveled indicator is not illuminated and the output power indication responds to the power control knob as it should.
It sorta puts out full unleveled power, but the actual output power varies from + 5 dbm to +15dbm depending on the frequency.
I originally thought this was an ALC fault (A4) and after much trouble shooting i noticed that after looking at the output with a spectrum analyzer i? that the output starting at 10mhz up until about 1 ghz consisted of multiple signals which slowly disappear as you approach 1 ghz.
After 1 ghz there is only the one signal at about 15-18 dbm ( i can only go to 1.8ghz on the spectrum analyzer).
Now for the tricky part, this distortion only occurs when actually sweeping the band, if i manually vary the frequency from 1.8ghz right down to 10mhz the output consists of one signal only!
Any suggestions as where to look?
Ken VK2TKE.
|
Hi Dave, ??????????????? The sweep rate for the sweeper was 10 seconds for 10 mhz to 1.8ghz as depicted in the picture ?
toggle quoted message
Show quoted text
From: [email protected] <[email protected]> On Behalf Of Dave Brown
Sent: Thursday, 20 February 2020 4:31 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] hp 83592A problem? What do you see when you slow the 8350 sweep rate down to be much slower than the sweep rate of the spec an? DaveB, NZ ? ? From: [email protected] [mailto:[email protected]] On Behalf Of Ken Goodhew
Sent: Thursday, February 20, 2020 18:04
To: [email protected]
Subject: [HP-Agilent-Keysight-equipment] hp 83592A problem ? Hi Gang,
I have an 8350b fitted with an 83592A plug in.
The problem exhibits as no control over the output power,but the unleveled indicator is not illuminated and the output power indication responds to the power control knob as it should.
It sorta puts out full unleveled power, but the actual output power varies from + 5 dbm to +15dbm depending on the frequency.
I originally thought this was an ALC fault (A4) and after much trouble shooting i noticed that after looking at the output with a spectrum analyzer i? that the output starting at 10mhz up until about 1 ghz consisted of multiple signals which slowly disappear as you approach 1 ghz.
After 1 ghz there is only the one signal at about 15-18 dbm ( i can only go to 1.8ghz on the spectrum analyzer).
Now for the tricky part, this distortion only occurs when actually sweeping the band, if i manually vary the frequency from 1.8ghz right down to 10mhz the output consists of one signal only!
Any suggestions as where to look?
Ken VK2TKE.
|
What do you see when you slow the 8350 sweep rate down to be much slower than the sweep rate of the spec an? DaveB, NZ ? ?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of Ken Goodhew
Sent: Thursday, February 20, 2020 18:04
To: [email protected]
Subject: [HP-Agilent-Keysight-equipment] hp 83592A problem ? Hi Gang,
I have an 8350b fitted with an 83592A plug in.
The problem exhibits as no control over the output power,but the unleveled indicator is not illuminated and the output power indication responds to the power control knob as it should.
It sorta puts out full unleveled power, but the actual output power varies from + 5 dbm to +15dbm depending on the frequency.
I originally thought this was an ALC fault (A4) and after much trouble shooting i noticed that after looking at the output with a spectrum analyzer i? that the output starting at 10mhz up until about 1 ghz consisted of multiple signals which slowly disappear as you approach 1 ghz.
After 1 ghz there is only the one signal at about 15-18 dbm ( i can only go to 1.8ghz on the spectrum analyzer).
Now for the tricky part, this distortion only occurs when actually sweeping the band, if i manually vary the frequency from 1.8ghz right down to 10mhz the output consists of one signal only!
Any suggestions as where to look?
Ken VK2TKE.
|
Hi Gang,
I have an 8350b fitted with an 83592A plug in.
The problem exhibits as no control over the output power,but the unleveled indicator is not illuminated and the output power indication responds to the power control knob as it should.
It sorta puts out full unleveled power, but the actual output power varies from + 5 dbm to +15dbm depending on the frequency.
I originally thought this was an ALC fault (A4) and after much trouble shooting i noticed that after looking at the output with a spectrum analyzer i? that the output starting at 10mhz up until about 1 ghz consisted of multiple signals which slowly disappear as you approach 1 ghz.
After 1 ghz there is only the one signal at about 15-18 dbm ( i can only go to 1.8ghz on the spectrum analyzer).
Now for the tricky part, this distortion only occurs when actually sweeping the band, if i manually vary the frequency from 1.8ghz right down to 10mhz the output consists of one signal only!
Any suggestions as where to look?
Ken VK2TKE.
|
Re: sick hp 5421A Digital Processor
What's the relation between the 5421A and the 5480A?? I have a volume from the 5480A service manual.
On Wed, Feb 19, 2020 at 9:50 PM Dave McGuire < mcguire@...> wrote: On 2/19/20 6:27 AM, ChrisBeee via Groups.Io wrote:
> Dear 5401A MCA afficionados,
> my beautiful 5401A has finally decided to quit the job, the 5421A
> Digital Processor stopped working. It started with a memory glitch /
> broken trace in the left half of the first (leftmost) sector, now there
> is only a fixed memory dump visible on the screen when reading the
> memory. Memory Erase does not work either anymore. I guess there is a
> problem in the digital section / state machine of the processor since
> the same symptoms showed with a second 5431A plug-in and the analog box
> seems to be functional. Any hint would be highly appreciated!
>
> Unfortunately, the original documentation seems to be made of
> unobtainium. Is there somebody out there who could help with a pdf or
> even an excess paper version of the SM? No question I would offer a
> proper compensation for the effort to ship a stack of paper over the
> pond, of course!
> It would be pretty cool if could get the 5401A competing against the
> (fully functional) 3721A again...
> Many thanks in advance!
? Wow, the HP MCA!? I'm sorry that I cannot offer you any documentation,
but I had to comment on this.? I knew it existed but I've never heard of
anyone having one.? Beautiful!
? And that's quite a wall of NIMs behind it.? Your lab looks a bit like
mine. =)? Have you done anything with CAMAC too?
? ? ? ? ? ? ? -Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
|
Re: sick hp 5421A Digital Processor
On 2/19/20 6:27 AM, ChrisBeee via Groups.Io wrote: Dear 5401A MCA afficionados,
my beautiful 5401A has finally decided to quit the job, the 5421A
Digital Processor stopped working. It started with a memory glitch /
broken trace in the left half of the first (leftmost) sector, now there
is only a fixed memory dump visible on the screen when reading the
memory. Memory Erase does not work either anymore. I guess there is a
problem in the digital section / state machine of the processor since
the same symptoms showed with a second 5431A plug-in and the analog box
seems to be functional. Any hint would be highly appreciated!
Unfortunately, the original documentation seems to be made of
unobtainium. Is there somebody out there who could help with a pdf or
even an excess paper version of the SM? No question I would offer a
proper compensation for the effort to ship a stack of paper over the
pond, of course!
It would be pretty cool if could get the 5401A competing against the
(fully functional) 3721A again...
Many thanks in advance! Wow, the HP MCA! I'm sorry that I cannot offer you any documentation, but I had to comment on this. I knew it existed but I've never heard of anyone having one. Beautiful! And that's quite a wall of NIMs behind it. Your lab looks a bit like mine. =) Have you done anything with CAMAC too? -Dave -- Dave McGuire, AK4HZ New Kensington, PA
|
Re: sick hp 5421A Digital Processor
Wow, I have never seen one of these. I did poke around in my usual sources, but also came up empty other than the 1969 catalog (). These were "price on request" items too...exceptionally expensive and low volume instruments I'm sure.
I do wish you success in solving the problem. Fingers crossed that someone on the list will have the service documentation squirreled away somewhere.
|
Re: sick hp 5421A Digital Processor
Hi Robert, actually, I do not do any nuclear stuff (yet ;-) ) with these MCAs (mainly because of the tight restrictions in my country), but audio noise characterization in the professional tape recording area. Interesting enough: hp obviously developed two more or less competing instruments at opposite ends of the world with similar functional specs, targeting at two wholly different markets (54xx series for nuclear, 3721A for signal analysis). I am only a humble restorer/user, but I honestly admire the ingenious brains behind these beautiful instruments!? Cheers Chris
|
Re: Replace floppy in 8720D with USB
toggle quoted message
Show quoted text
On Feb 19, 2020, at 8:00 AM, Pete Manfre <pmanfre@...> wrote:
? I'll send you some?
Pete wa2odo? On Wed, Feb 19, 2020, 10:57 AM Dave Miller < ve7hr@...> wrote: Being in Canada that link did not seem to work but I think I found it on the US site. I will give it a go.? I also on the Canada Amazon found a really expensive one that might fit. It was abou $200 USD. I will try cheap one first.?
This should be interesting.? I appreciate all the help.? Floppies seems much simpler at this time. I will have to find one.? Purged my thousand of them about 15 years ago.? Dave VE7HR? On Feb 19, 2020, at 12:34 AM, pianovt via Groups.Io <pianovt=[email protected]> wrote:
?Here is a Chinese seller on Amazon. The odd thing is, it says "1 lot = 5 pieces". I guess, that means you get five for that price. Or, maybe not. Anyway, I brought this up a speculative guess, so "buyer beware".
Vladan
|
On 2/19/20 1:20 PM, Askild wrote: If somebody made an open source android app for the seek camera, that
would be amazing!
I don't have the knowledge to do it, but I'm sure there are some that
can do it, with all the info from the pc software available..
And I would probably soon buy me a camera... Someone did write an open-source Linux program for displaying video from the Seek cameras. I've used it; it works. In fact, now that Seek is being a bunch of suitly dickheads, that's the only way I use mine. -Dave -- Dave McGuire, AK4HZ New Kensington, PA
|
If somebody made an open source android app for the seek camera, that would be amazing!
I don't have the knowledge to do it, but I'm sure there are some that can do it, with all the info from the pc software available.. And I would probably soon buy me a camera...
Br, Askild
On Wed, Feb 19, 2020 at 2:54 PM Chuck Harris < cfharris@...> wrote: I have just been in a long conversation with SEEK technical support,
and they claim their camera must have this information to allow you
to speak notes, and leave watermarks of location and other information
on every picture... Yeah right!
Further, scanning through the hundreds of bad reviews for SEEK's AP
on playstore, there are claims that the AP wakes up periodically, and
reads the GPS, and goes onto the internet...
My early version of the AP didn't enforce the requirements of location
and microphone, but did periodically stop the camera, and drop me into
an advertisement screen.? The only way back was to restart the AP.
So, I guess it is time to make an open source AP for Android and quit
this nonsense.
One thought that occurs to me:
Advanced FLIR like the SEEK XR camera is considered to be export
restricted munitions, according to the US State Department.
Perhaps the geolocation and internet connection is being required by
the US State Department?
A company would have to be suicidal to treat their customers the way
SEEK appears to be treating us.
-Chuck Harris
Andrew Hakman wrote:
> Wow, nice, so open source software solves the problem of the company being a turd and
> wanting to spy on you! Sweet.
>
> Might have to be on the lookout for a cheap used thermal camera that other people are
> ditching because the app sucks!
>
> On Tue, Feb 18, 2020 at 9:17 AM Razvan <razvan@...
> <mailto:razvan@...>> wrote:
>
>? ? ?Hello Askild,
>
>? ? ?I'm replying to the public thread, because I got other requests for this, so it's
>? ? ?probably better if I just put the links here:
>
>? ? ?GitHub:
>? ? ?Binaries:
>
>? ? ?Keep in mind I lost interest in this piece of software, so I won't consider
>? ? ?making changes to it. Several other people have made use of my algorithm in their
>? ? ?own programs, so there should be other alternatives out there, featuring similar
>? ? ?image quality (just search the EEVBlog).
>
>? ? ?Best regards,
>? ? ?Razvan
>
>? ? ?Razvan Cojocariu
>? ? ?
|
Dave McGuire
