HP8673C Restoration project
Hello everyone,
?
My first post. I acquired a HP8673C 18 GHz synthesizer at an auction when a company
called Plessey collapsed locally many years ago. I was working there as an engineer at the
time using this equipment.
?
I have fixed various issues like the smoking capacitor across the mains input, replaced the
bottom fan with a more modern variant and one of the supply rails needed repairing.
?
I am now at a point where I have an unlocked failure mode on the main YTO loop on Service
Sheet BD4 (the light is off). The YTO is generating signals, but off frequency.
I have arrived at point 3 on the DAC Check where I an not getting the -6V drive signal at 2GHz.
?
My problems are these :
1: I seem to have a later variant of the A3A5 DAC board than is described in the commonly available
service manuals on the internet. Off course none of the mod status or change notes are available
online. There does seem to be a mention of this newer board on the service manual on the Keysight
site (Change 25) , but it of such a poor scan quality to basically render it useless.
2: To proceed to the next step I will need to acquire a 36 pin riser card. I expect this is a common
problem and I am sure there are sources available for these.
?
If anyone can point me in a direction to obtain a better quality of the most recent copy of sheet BD4
and the extender cards that would be greatly appreciated.
?
73's
Allan / Cape Town / South Africa.
?
?
|
Thanks, guys; some very useful info overnight! I should have mentioned the analyzer I'm using is the good old venerable 8753C. Though its screen is growing somewhat dim, it still does the job. I just have to close the curtains if it's too bright outside. :-)?
|
Re: Service manual scan post processing
High Peter,
I'm using "ImageOptim". I think its free. It works on JPEG and other formats and has a number of compression algorithms where it chooses the one that fits most. You can decide if you accept some loss of information or not.
Usually, taking the pics of my camera that come in at 3-4MB (medium size setting on the camera, whatever that means), they are reduced by 90% on average without me seeing any loss of information.
cheers
Martin
|
Disassembling HP 64000 code
Is anyone here familiar with data structures built by the HP 64000 system? Specifically "6809 79-STD FORTH w/HP64000 simulated I/O ?BJS Rev. 09/01/83" While disassembling firmware code for the HP 3457A multimeter I found a block of code that is inaccessible from the code while running in the meter. It appears to be symbol tables and simulator code for the 64000 build environment. I was able to use it to give names to all of the functions but there are some parts of the data structure I can't figure out. There were also some references to FORTH but I couldn't find any info about FORTH being a supported language on the 64000. Pascal Strings are used widely in this code but they are slightly different. While the strings follow the form of a one byte pascal string, with a byte value preceding the string and describing the length of the string, they often repurpose some bits in the leading and trailing byte. For example, of the maximum string length is less than 32 characters (like for display strings) the lower 5 bits of the first byte are used for length and the upper 3 bits are flags for something. Sometimes, like in a link list, the flags describe if there is data following the string which might be code, pseudo code, or address to code. Or maybe if the entry is the first or last in the list. The final byte is the last character of the string but since ascii only needs 7 buys, the 8th bit is often used as a flag. Are these oddities something built into the 64000 compiler and used for debug code or is it likely a 3457A specific implementation?
|
Re: Dissassembling HP 65000 code
Sven, you are right. It is the 64000 and FORTH although Pascal strings play a large part. I was being lazy and insted of checking the disassembled code to verify my information I relied on my memory. It is a lesson I have to relearn far too often. I will restart the thread with the correct information for posterity.
"Tom Haynes via " <=[email protected]> writes:
> Is anyone here familiar with data structures build by the HP 65000
> system?
Are you sure it's 65000? I only know about the 64000 and 647000 systems.
My firmware has:
00001b60? 42 1e a8 61 f6 1c fa 60? 42 12 5c 10 42 3c 36 38? |B..a...`B.\.B<68|
00001b70? 30 39 20 37 39 2d 53 54? 44 20 46 4f 52 54 48 20? |09 79-STD FORTH |
00001b80? 77 2f 48 50 36 34 30 30? 30 20 73 69 6d 75 6c 61? |w/HP64000 simula|
00001b90? 74 65 64 20 49 2f 4f 20? 20 42 4a 53 20 52 65 76? |ted I/O? BJS Rev|
00001ba0? 2e 20 30 39 2f 30 31 2f? 38 33 22 68 10 42 3c 4d? |. 09/01/83"h.B<M|
00001bb0? 6f 64 69 66 69 65 64 20? 66 6f 72 20 52 4f 4d 20? |odified for ROM |
00001bc0? 62 61 73 65 64 20 6f 70? 65 72 61 74 69 6f 6e 20? |based operation |
> While dissassembling firmware code for the HP 3457A multimeter I found
> a block of code that is inaccessible from the code
> while run in the meter. It appears to be symbol tables and simulator
> code for the 65000 build environment. I was able to
> use it to give names to all of the functions but there are some parts
> of the data structure I can't figure out.
>
> There was also some references to PASCAL but I couldn't find any info
> about PASCAL being a supported language on the
> 65000. Maybe it was just describing the use of Pascal Strings.
Look on , especially the files in
It contains the full source of HP's version of pascal. I only know it
from HP 9000/300 workstations, but i wouldn't be surprised if it was
also used for/in other projects. On the other hand does the comment
above say FORTH, but i have no idea about FORTH, so can't say anything
about that.
BTW, i hacked a pseudo driver for the 3457a years ago in mame to do some
reverse engineering. Not sure if it's of any help, but here's the commit:
It's good enough to display an Error message on the emulated display
because it can't talk to any measurement hardware.
|
Re: Service manual scan post processing
Peter,
I have scanned quite a few manuals. I will take a look at the scans. I find it is always best to ingest the scans as tif (lossless compressed with lzw or something) and use a separate tool to convert to pdf as one of the final steps in processing. As a rule I never use the scanner's scanning software beyond what is absolutely necessary.
Never successfully scanned a fiche, though. Have a Minolta MS-6000 with a broken SCSI interface. The MS-800 looks nice. Have been eyeing a ScanPro 2000/3000 for fiche for years but never found the right auction.
As for OCR I use Tesseract. Free, very good, and easy to use.
A tool like tiff2pdf like Mr. Gardner pointed out works well. I think my script uses img2pdf which is probably similar.
-Michael Bierlein
On Mon, Feb 17, 2025 at 09:48 Peter Brown < peter@...> wrote: If anyone wants to have a go, there are some sample scans here?
?
Files - A temporary directory for photographs and help relating to emails and posting - 8430A 08340-90021 Service vol 1 section1
?
They are PDF'd from the scanning software with minimum compression.? The software will also produce .tiff files if these are a better place to start
?
Peter
|
Re: Dissassembling HP 65000 code
"Tom Haynes via groups.io" <tomhaynes.net@...> writes: Is anyone here familiar with data structures build by the HP 65000
system? Are you sure it's 65000? I only know about the 64000 and 647000 systems. My firmware has: 00001b60 42 1e a8 61 f6 1c fa 60 42 12 5c 10 42 3c 36 38 |B..a...`B.\.B<68| 00001b70 30 39 20 37 39 2d 53 54 44 20 46 4f 52 54 48 20 |09 79-STD FORTH | 00001b80 77 2f 48 50 36 34 30 30 30 20 73 69 6d 75 6c 61 |w/HP64000 simula| 00001b90 74 65 64 20 49 2f 4f 20 20 42 4a 53 20 52 65 76 |ted I/O BJS Rev| 00001ba0 2e 20 30 39 2f 30 31 2f 38 33 22 68 10 42 3c 4d |. 09/01/83"h.B<M| 00001bb0 6f 64 69 66 69 65 64 20 66 6f 72 20 52 4f 4d 20 |odified for ROM | 00001bc0 62 61 73 65 64 20 6f 70 65 72 61 74 69 6f 6e 20 |based operation | While dissassembling firmware code for the HP 3457A multimeter I found
a block of code that is inaccessible from the code
while run in the meter. It appears to be symbol tables and simulator
code for the 65000 build environment. I was able to
use it to give names to all of the functions but there are some parts
of the data structure I can't figure out.
There was also some references to PASCAL but I couldn't find any info
about PASCAL being a supported language on the
65000. Maybe it was just describing the use of Pascal Strings. Look on bitsavers.org, especially the files in It contains the full source of HP's version of pascal. I only know it from HP 9000/300 workstations, but i wouldn't be surprised if it was also used for/in other projects. On the other hand does the comment above say FORTH, but i have no idea about FORTH, so can't say anything about that. BTW, i hacked a pseudo driver for the 3457a years ago in mame to do some reverse engineering. Not sure if it's of any help, but here's the commit: It's good enough to display an Error message on the emulated display because it can't talk to any measurement hardware.
|
Dissassembling HP 65000 code
Is anyone here familiar with data structures build by the HP 65000 system?
While dissassembling firmware code for the HP 3457A multimeter I found a block of code that is inaccessible from the code while run in the meter. It appears to be symbol tables and simulator code for the 65000 build environment. I was able to use it to give names to all of the functions but there are some parts of the data structure I can't figure out.
There was also some references to PASCAL but I couldn't find any info about PASCAL being a supported language on the 65000. Maybe it was just describing the use of Pascal Strings.
On the topic of Pascal Strings, they are widely used in this code but they are slightly different. While the strings follow the form of a one byte pascal string, with a byte value preceding the string and describing the length of the string, they often repurpose some bits in the leading and trailing byte. For example, of the maximum string length is less than 32 characters (like for display strings) the lower 5 bits of the first byte are used for length and the upper 3 bits are flags for something. Sometimes, like in a link list, the flags describe if there is data following the string which might be code, pseudo code, or address to code. Or maybe if the entry is the first or last in the list. The final byte is the last character of the string but since ascii only needs 7 buys, the 8th bit is often used as a flag.
Are these oddities something built in to the 65000 compiler and used for debug code or is it likely a 3457A specific implementation?
|
That's an interesting setup. I would start with a two-port S21 Response calibration.
?
Not sure what VNA you're using. My 8510+8515A won't go that low of frequency but if yours are similar to 8510,
it's?
1) press CAL
2) Select CAL kit(I think any calkit will do for your setup)
3) Press S21,
4) Select "Calibrate Response"
5) Connect thru
6) Press "Thru" standard key.
7) Done and save CAL Set#
?
Then measure S21.?
?
Good luck!
Calvin
?
|
You can rest assured that none of those are for 50 ohms or even close, but you can still fool around with them on a VNA and get some ideas of what they are.
?
Those color codes can identify them somewhat since the functions of the ones used in transistor radios were somewhat standardized long ago and assigned to color codes, but different manufacturers may have their own or modified codes. But, it's a good place to start. A good info source is the fun with tubes group here:
?
https://funwithtubes.groups.io/g/main/messages
?
I know I have seen and been involved in some pertinent IF transformer discussions there, and have seen references to some sites with info on the color codes and such, within the past couple years or so. It's not a high traffic group, so it should be fairly easy to search for the right stuff there. It's easy to join if necessary to ask for advice and look through their archives.
?
I have similar but much larger troubles with IF cans of all sorts, with small numbers of very many kinds that I have pulled and saved over decades - lots of brands, sizes, styles, color codes, and so on. I started sorting them somewhat a couple years ago, but have a very long way to go, and most will never be properly ID'd. I have maybe a thousand pieces. The ones you have probably were high volume broadcast radio units with standard functions in the signal chains, seven different ones may have been all that were needed for a wide range of AM/FM transistor sets.?
?
The easiest to spot are the IFTs, which often have a tiny dogbone ceramic capacitor mounted in a cavity on the bottom. These tune for the nominal IF 455 kHz AM or 10.7 MHz FM and possibly 19/38 kHz pilots for stereo decoding. Sometimes AM LO and FM discriminator transformers have the built in cap too, which can confuse things a bit. I think FM LOs usually used open coils that are deformed for tuning so not in cans. The cans without the caps are more general purpose. It's possible to have TV parts too, including video IF around 43 MHz VSB, and audio IF and traps around 4.5 MHz. My collection is mostly in these types but includes lots of others like VCRs, cordless and cell phones, ISM bands, CB radios, satellite, cable, and industrial stuff. I also have a lot of just plain variable inductors, which are very handy, and of course, using only one winding of an appropriate transformer can serve the purpose. Note also that many types can be opened up and modified.?
?
Another big clue is the style of the magnetic tuning slug. If it's fat (and will be ferrite) and goes around the outside of the coil assembly like an inverted cup, it's likely for very low frequencies below 10 MHz. That yellow one in your? picture might be that kind. Try to screw it all the way down and see if the coil form is visible under the can lip. If not, it's a low frequency one. Midband types will have more of a screw-shaped slug of ferrite (lower end) or powdered iron (upper end) that threads into the coil form. If the slug is aluminum, that's for VHF like satellite 480 or 600 MHz IF and helical filters.
?
Anyway, the nature of the part will tend to show on a VNA even without matching - it's a sloppy measurement, but probably enough to ID the basic function types. The best thing to start with is a 50 ohm series damping R in each line to/from the VNA which will reduce any crazy resonances so things are a little easier to decipher. No fancy fittings are needed - just tack in whatever parts help get a cleaner picture. It's actually a helluva lot of fun to experiment with and re-task various IF can type parts into new uses. I have done it a lot especially for custom LC filters, where adjustable tuning is essential. I like to use fixed padded mica caps with variable slug-tuned chokes in cans for most filters below a hundred MHz.
?
Ed
?
|
Re: HP 8563E Shows Signal Present at 100MHz and below without an Input
What marker frequency do you get on the 0dBm signal?
I am currently not experiencing the same issue today, yesterday when it occurred the peak search would not find the phantom signal.
It sounds like you are looking at the LO feed through signal, if so the marker on peak signal will be 0 MHz @ 0 dBm.?
I am now experiencing a large phantom spike at 0Hz at 1.67dBm.?
What do you see if you do an instrument preset (green button)?
?
Last State as expected. The Span is 26.5GHz the Start is 0Hz the Stop is 26.5GHz and the Center is 13.25GHz.
At turn on, what is the start frequency? 0Hz What is the stop frequency? 26.5GHz?
?
With cal signal connected, can you run the self cal procedure? I am not sure where the "Self cal" procedure is located. When pressing the "Cal" button and entering the Cal menu I am able to "Realign th eLO and IF, turn on or off the IF Adj, Adjust the current IF state, perform a full IF adjust, perform a Ref level adjust and a few item on the second page.?
?
What happens when you narrow the span to 100, 10, 5.01, 5.00, 1 MHz, and 100KHz? Span appears to be working. Do it with the phantom signal, then with the cal signal. Cal signal 300 MHz @ -10 dBm
Please report back your findings.
?
Thank you for your assistance!
|
Searching parts for HP 5370B
Hello to all members.
I need mechanical parts to repair my HP 5370B. Sadly the parts list in the service manual I found on the web is not too good, so I don't know the part numbers.
- display filter
- front panel frame
Please PM if you have these parts available.
Thank you !
Claude VA2 HDD
?
|
Re: Fan and ball bearings on 8340/8341 Sweep Generators
You can always get replacement bearings for fans like this.?
toggle quoted message
Show quoted text
On Feb 17, 2025, at 4:31?PM, Calvin Guan via groups.io <guancalvin@...> wrote:
? Thanks for all replies.
?
Hi Jim,
?
I followed your advice to pry it open and sure enough, there is no hope to put the seal back on but as you said that's no necessary.
As the two bearings with remaining seal will form a close chamber.
?
Now the unit is as quiet or laud as other 8340s which makes me happy.
?
What modern fan is good for 8340/8341? I haven't found one yet.
?
Calvin
?
?
<pry_it_open.jpg> <Old_grease.jpg> <clean_up.jpg> <new_grease.jpg> <Install inboard.jpg> <install outboard.jpg> <Make Fan Great Again.jpg>
|
Re: Fan and ball bearings on 8340/8341 Sweep Generators
Thanks for all replies.
?
Hi Jim,
?
I followed your advice to pry it open and sure enough, there is no hope to put the seal back on but as you said that's no necessary.
As the two bearings with remaining seal will form a close chamber.
?
Now the unit is as quiet or laud as other 8340s which makes me happy.
?
What modern fan is good for 8340/8341? I haven't found one yet.
?
Calvin
?
?
|
Re: HP 3336B series 2902 repair
Hello to all members.
I found out that a service note might be part of the answer, but so far I was not able to find it, it is service note 3325-A21.
Thank you and best regards,
Claude
|
|
|
Perhaps it's because it's very late here in my timezone, but I'm not really seeing why the 50 ohm system impedance would be an issue in this regard?
I'll have to read any answers to that tomorrow as it's past time for me to hit the hay.
If there are any other issues, please summarize them so I can address them at the same time as the above.
?
thanks,
?
J.
|
Re: Looking for RF Limiter specs
Don, thank you. I have no trouble determining frequency range desired, and have most of the necessary part numbers for the specific instruments.? I've been concerned about how to substitute similar spec'd parts should I not be able to locate direct replacements. Also, I need complete specs to be able to test what I do find...
?I think I've only looked in one or two accessory catalogs, unsuccessfully. I foolishly figured they'd all be the same. Maybe the answers were just sitting there waiting for me had I been more thorough.
?
Thanks again.
|
You can make wideband transformers with small toroids and a bit of wire.?
I did that to test some crystal filters that were closer to 500 ohms. You can also make resistive pads.?
At 50 ohms the response looked horrible and I couldn’t even identify the BW but with the transformers the response looked quite proper.?
toggle quoted message
Show quoted text
On Feb 17, 2025, at 3:15?PM, Jinxie via groups.io <paul666@...> wrote:
? The impedance of the transformers obviously changes with frequency. They are, in effect, band-pass filters. As for the 50 ohms, that's the 'system impedance' and there's nothing I can do to change that!
I did remember *after* I'd gone to all this trouble that I do actually have a vintage 'wobbulator' in good working order which would have been a much more sensible approach than using a VNA!
|
The impedance of the transformers obviously changes with frequency. They are, in effect, band-pass filters. As for the 50 ohms, that's the 'system impedance' and there's nothing I can do to change that!
I did remember *after* I'd gone to all this trouble that I do actually have a vintage 'wobbulator' in good working order which would have been a much more sensible approach than using a VNA!
|
Jinxie
Martin
Pete_G4GJL
Dave Borcher