UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Test Gear and Workshop Equipment

Notices

Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders.

Reply
 
Thread Tools
Old 19th May 2020, 12:41 am   #101
Julesomega
Heptode
 
Julesomega's Avatar
 
Join Date: Dec 2012
Location: Stockport, Greater Manchester, UK.
Posts: 688
Default Re: HP8640B Signal Generator

I'm so impressed, yours was in the worst possible condition so for anyone else it looks relatively simple to keep an 8640B working, and to specification.

Now all you need is a down-converter to cover the LF spectrum! I'll scan some documents about this asap
__________________
- Julian

It's good here
Julesomega is offline   Reply With Quote
Old 19th May 2020, 12:54 am   #102
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 208
Default Re: HP8640B Signal Generator

The battle isn't over yet Jules. The decimal point circuit isn't working properly and it's years since I tackled 74 series logic. I suspect a chip has failed and it's not a very common one. The good news is I found one for 99p so I should know in a few days whether I correctly diagnosed a duff 74LS96 or not.
I also found a cracked gear hidden away on the modulation frequency control.

OK on the downconverter. I should be able to make a simple one with a suitable crystal and diode ring perhaps? It'll make an interesting little project.
cheers
Allan
allan is offline   Reply With Quote
Old 19th May 2020, 1:32 pm   #103
Nymrod121
Octode
 
Nymrod121's Avatar
 
Join Date: Apr 2012
Location: Worcestershire, UK.
Posts: 1,786
Default Re: HP8640B Signal Generator

Quote:
Originally Posted by Julesomega View Post
(snipped)

Now all you need is a down-converter to cover the LF spectrum! I'll scan some documents about this asap
That sounds like a useful option - I'd like to give it a go, if I may?
Best wishes
Guy


(PS: I'm still planning on a visit to Manchester re. the Tek transformer you kindly kept for me ... by Covid-19 ...)
__________________
"Halt! - who goes there?"
'Baa-aa-aah ..." ('The Dish': 2000)
Nymrod121 is offline   Reply With Quote
Old 19th May 2020, 8:15 pm   #104
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 208
Default Re: HP8640B Signal Generator

Decimal point is on the way... when the new LS74LS96N arrives and the pesky A8A3 is soldered back, coax connected and casing replaced.
Let's hope it's the only duff chip because some are as scarce as hen's teeth.
Allan
Attached Thumbnails
Click image for larger version

Name:	leaper.jpg
Views:	39
Size:	42.8 KB
ID:	206245  
allan is offline   Reply With Quote
Old 20th May 2020, 9:22 am   #105
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 208
Default Re: HP8640B Signal Generator

Quote:
Originally Posted by allan View Post
OK on the downconverter. I should be able to make a simple one with a suitable crystal and diode ring perhaps? It'll make an interesting little project.
cheers
Allan
Here's a very simple way of extending the range of the HP8640B down to DC.
I used the same circuit in a home brew upconverter for a Lime SDR with poor HF performance.
The HP provides a 5MHz output at around 100mV.
Maybe a few frills would improve it, for example a couple of resistors to make the output level reflect the numbers for attenuator calibration by -10.
Housed in a small diecast box in which is mounted a small piece of tin with parts soldered in place. No power supply needed.
Allan
Attached Thumbnails
Click image for larger version

Name:	downconverter.jpg
Views:	43
Size:	34.1 KB
ID:	206260  
allan is offline   Reply With Quote
Old 20th May 2020, 10:19 am   #106
Radio Wrangler
Dekatron
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 14,531
Default Re: HP8640B Signal Generator

From the dirty tricks department:

You can extend the bandwidth of those two transformers by changing the single winding to another centre tapped bifilar one.

You wind the transformer with quadrafilar wire.... four insulated strands twisted together.

Ground the centre tap, apply/take signal at one end and let the other end float.

Sounds barmy?

The winding to nothing has an antiphase signal along it compared to the winding being used and this corrects a capacitive imbalance between the windings which is inherent in the 3-winding version. One of the transmission line transformer structures Ruthroff missed!

HP sold their mixer diodes as either singles or as matched quads. No longer available now. HP semiconductors became Broadcom then dropped all their useful parts.
BUT a diode test on a dvm hits the diode with a fixed current and measures the voltage drop. This is good enough for basic matching, just find the closest four you've got.

A few basic tricks can tart up a plain mixer quite handily.

Oh, you can also use indelible markers to colour your strands differently, just pull the wire over the felt nib. Saves a few minutes juggling with ohmmeters to find the various ends.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline   Reply With Quote
Old 20th May 2020, 10:52 am   #107
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 208
Default Re: HP8640B Signal Generator

The Broadcom ring is obsolete David, but applying the well known trick of using the longest possible code for the smallest part, you can use a
BAT15099RE6327HTSA1 (Farnell @ 24p +vat) or if you're an RS account holder they have them for 25p each but you'd have 2,999 in your junk box and be 864 poorer (not to mention having to wait till mid-July.
Allan
allan is offline   Reply With Quote
Old 20th May 2020, 11:23 am   #108
Radio Wrangler
Dekatron
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 14,531
Default Re: HP8640B Signal Generator

Aye, I'm familiar with all that. Whenever RS send someone round trying to drum up some more business, they do tend to get beaten up.

If bits for prototyping aren't affordable, then they aren't going to get designed in, are they.

All the distributors went through a phase where they decided that their real customers were the purchasers buying production quantities, so they put max effort into wooing them. All importance was removed from the concept of design engineers wanting data in a form where they can choose between parts easily and where they can buy tiddly components a few at a time for small change. They haven't sussed out the small inconvenience that if they make it hard for the designer, the corporate purchasing people will be looking for different parts from different distributors. Success requires everyone to be able to do their jobs.

Half that code probably says it's lead free and not made of antimatter.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline   Reply With Quote
Old 20th May 2020, 12:22 pm   #109
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 208
Default Re: HP8640B Signal Generator

My favourite source is CPC Bargain Basement David.
Using F3 to search, Dual Schottky diodes for 2p today.
I might rustle up a quick downconverter if I can locate my spare ring diodes chips while waiting for the 74LS96.
I thought an MD108 and dispense with the transformers, but I might need a spot more dB of 5MHz and that would negate the zero power supply?
I did find a few ancient HP circuit boards yesterday which probably carry some useful bits and pieces eg there are a few of those top hat rectifiers.
Allan
allan is offline   Reply With Quote
Old 20th May 2020, 12:31 pm   #110
G0HZU_JMR
Nonode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 2,056
Default Re: HP8640B Signal Generator

There is an old HP application note covering this, see the attached pdf below.

I made something similar many years ago for my old Marconi TF2015 signal generator to extend the range below the 10MHz lower limit of the TF2015.

I used a 30MHz oscillator and I also included a gain stage to overcome the loss in the mixer. This meant that the output level remained the same as the sig gen amplitude setting.

Note that if you do make a downconverter, you can then convert down to a signal within the range of a PC soundcard. The log accuracy of the soundcard should be extremely good so it should be possible to check the accuracy of the (rotary) vernier attenuator on your HP8640B within this frequency range. This will give an initial indication as to how suitable the IV characteristic of your detector diode is. This will only apply at low test frequencies of a few MHz but you could do the same at higher frequencies with a suitable mixer and another sig gen to provide the (VHF/UHF?) LO for the mixer.
Attached Files
File Type: pdf an_171-2.pdf (1.06 MB, 21 views)
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline   Reply With Quote
Old 20th May 2020, 12:53 pm   #111
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 208
Default Re: HP8640B Signal Generator

I'd spotted that note Jeremy.

I've had a TF2008 for many years.
It goes down to zero frequency and has a very clever end-to-end tuning arrangement from zero to over 500MHz. I use it when I need to find a response 100 times faster than my Wavetek 2407.
My upconverter is here.
http://www.radiomuseum.co.uk/converter.html#anchor19162
Not a wonderful design, but worked really well for the job in hand.
Allan
allan is offline   Reply With Quote
Old 20th May 2020, 2:25 pm   #112
Julesomega
Heptode
 
Julesomega's Avatar
 
Join Date: Dec 2012
Location: Stockport, Greater Manchester, UK.
Posts: 688
Default Re: HP8640B Signal Generator

That's the app note I was thinking of Jeremy, can't improve on that
The matching HP 11710B Downconverter can be seen on Radiomuseum
If anyone wants any of those BAT15 series just name your pinout, I think I've got them all

P.S. Anyone got an unwanted HP ring binder so I can keep the three manuals and the app note together and accessible?
__________________
- Julian

It's good here
Julesomega is offline   Reply With Quote
Old 20th May 2020, 11:07 pm   #113
Julesomega
Heptode
 
Julesomega's Avatar
 
Join Date: Dec 2012
Location: Stockport, Greater Manchester, UK.
Posts: 688
Default Re: HP8640B Signal Generator

Here for the archive are technical details of the HP 11710B Downconverter
Attached Files
File Type: pdf HP 11710B.PDF (1.63 MB, 16 views)
__________________
- Julian

It's good here
Julesomega is offline   Reply With Quote
Old 23rd May 2020, 12:49 pm   #114
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 208
Default Re: HP8640B Signal Generator

Today, after about a week of waiting, my new 74LS96 chip (made in 1982!) arrived and my tester said Pass.
Some soldering and manipulating got the two circuit boards back in place and switching on I could see the decimal place was OK.
Connecting a 100MHz scope proved the outputs were a bit odd because the shaft between the range switch and the filter cam box was out by several steps.
Once I'd corrected this the ranges 500KHz to 64MHz shown on the counter corresponded to RF outputs having the same frequencies.
But.. the following ranges were wrong.
128 read 1.000, 256 read 2.000 and 512 read 4.000
The scope did manage to see 128MHz so I guess the outputs are OK.

I could set the meter to read 0dBm and it was constant from 512MHz down to 4MHz.
Below this 2MHz read -5dBm max and 1.0MHz read about -15dBm and 0.5MHz hardly anything but the RF output was fine. Perhaps a 10nF in place of the 33pF in the home brew RF sniffer will correct this?

RF output-wise the scope gave me 9.5V RMS to 13.4V RMS from 512MHz down to 0.5MHz. I used 50 ohm coax between the sig gen and the scope.

Some more fault finding is required. For starters, this is between the counter circuit and its input. Maybe a pre-scaler is involved but I haven't looked yet.
Feeding an external UHF signal into the counter should help?
Allan
Attached Thumbnails
Click image for larger version

Name:	leaper6.jpg
Views:	23
Size:	37.1 KB
ID:	206523   Click image for larger version

Name:	74ls96.jpg
Views:	29
Size:	69.1 KB
ID:	206524  
allan is offline   Reply With Quote
Old 23rd May 2020, 6:01 pm   #115
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 208
Default Re: HP8640B Signal Generator

Fixed... the RF output was OK but the counter display was wrong due to a fault in the logic used to set the correct divide function. Read the end of the page.
At first I thought I'd have to dismantle one of the mode switches.
http://www.radiomuseum.co.uk/HP8640B.html
Allan
allan is offline   Reply With Quote
Old 24th May 2020, 7:09 pm   #116
G0HZU_JMR
Nonode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 2,056
Default Re: HP8640B Signal Generator

Good work! Does the rotary vernier adjust the output power seen on the RF Level meter? You must be fairly close to having a fully working HP8640!

There's some interesting info about the levelling system in the HP app note attached below. This only covers operation up to about 512MHz and I guess things become more challenging if the generator has the 1024MHz frequency doubler option fitted.
Attached Files
File Type: pdf an170.pdf (2.31 MB, 14 views)
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline   Reply With Quote
Old 24th May 2020, 8:20 pm   #117
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 208
Default Re: HP8640B Signal Generator

Yes, the vernier works perfectly. That was the pot on the rear of the attenuator that I changed.
I'm not too happy with the RF output on the low ranges and I need to work through all the meter adjustment pots and maybe bridge the 30pF on my RF detector with say 2.2nF or something greater. More experimenting needed.

Today I connected it to a spectrum analyser and realised the shortcomings re harmonics although only the ranges 500KHz to 4MHz are a problem.
I don't have the doubler fitted, in fact I don't really have a requirement up in those regions so won't miss it. I do have Option 001.

Signal generator output voltage is always a puzzle to me. Is the figure on the meter the open circuit voltage and, if connected to an external 50 ohm load, is this recorded externally as half that on the attenuator dial, and proportionally the number of dBm. Is the level indicated on the meter different to measured output power on a 50 ohm spectrum analyser?
The meter read 0dBm and the SA read +6.54dBm.. Then again I haven't gone through the calibration exercise so 0dBm on the meter could well be wrong...I suppose the handbook will cover this?

I haven't confirmed all the modulation bits and bobs work yet, but I did check wbfm on a broadcast receiver in passing and it worked OK. In fact when I tuned off a local FM broadcast to check it, a voice in Spanish came up... must be the weather, or maybe global warming?
Allan
Attached Thumbnails
Click image for larger version

Name:	256Ma.jpg
Views:	22
Size:	40.9 KB
ID:	206682   Click image for larger version

Name:	256Mb.jpg
Views:	24
Size:	51.1 KB
ID:	206683  
allan is offline   Reply With Quote
Old 24th May 2020, 8:56 pm   #118
G0HZU_JMR
Nonode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 2,056
Default Re: HP8640B Signal Generator

I haven't used an HP8640 in over 30 years but looking at the hi res image in the link below I think the voltage on the meter is Vpd. This is the voltage seen at the output connector when the sig gen is correctly terminated externally with a 50R load.

http://hpmemoryproject.org/pict/time...40b_frt_hr.jpg


The image shows 0dBm and about 225mV (Vpd) on the dial on the 0-3 scale range and this agrees with the above. 0dBm is about 224mVrms across a 50R load.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline   Reply With Quote
Old 24th May 2020, 9:33 pm   #119
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 208
Default Re: HP8640B Signal Generator

That picture shows something amiss.
Did you spot it?
Allan
allan is offline   Reply With Quote
Old 24th May 2020, 9:43 pm   #120
G0HZU_JMR
Nonode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 2,056
Default Re: HP8640B Signal Generator

The frequency range selector seems to disagree with the 28MHz on the display?
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline   Reply With Quote
Reply

Thread Tools



All times are GMT +1. The time now is 12:32 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2020, vBulletin Solutions, Inc.
Copyright ©2002 - 2020, Paul Stenning.