UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV 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.

Closed Thread
 
Thread Tools
Old 2nd May 2020, 5:41 pm   #21
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
Default Re: HP8640B Signal Generator

As regards the allen key sizes, 1.27mm is exactly 0.050" (50 thou). In fact I think the smallest 3 (common) imperial Allen keys (0.028", 0.035", 0.050") are the same as the smallest 3 metric allen keys (even if some companies sell both imperial and metric versions under different part numbers!)
TonyDuell is online now  
Old 2nd May 2020, 8:36 pm   #22
factory
Octode
 
Join Date: Jul 2009
Location: Cheshire, UK.
Posts: 1,706
Default Re: HP8640B Signal Generator

Quote:
Originally Posted by Jon_G4MDC View Post
Really lovely generators. It is just a shame about that plastic which goes brittle.

Has anyone come up with 3D printed replacement parts yet?
Quote:
Originally Posted by allan View Post
I repaired a CNY2 receiver some time back and had to source some new gears.
It seems there's a lot available quite cheaply once you work out how they're specified, I imagine the gears in the 8640 have standard parameters.. maybe not though?
Allan
The gears that split in these are a combination of a straight & bevelled gear. They fail for the same reason as the knobs do, due to the centre being made from brass, the plastic outer section splits at the weak point where the grub screws are.
Click image for larger version

Name:	SAM_0632a.jpg
Views:	139
Size:	86.1 KB
ID:	204640

I took the easy solution & ordered the brass gears that an ebay seller was making, only problem is I haven't got round to fitting them yet. I did think about getting standard brass gears & brazing them together, but I seem to remember the cost was similar or more than the ready made set.
Note: they cost more than your 8640 including shipping & fees from India, ebay listing https://www.ebay.co.uk/itm/153024403695, the seller is currently away, it's probably best to fix the other problems first anyway.
Click image for larger version

Name:	SAM_1628a.jpg
Views:	159
Size:	87.1 KB
ID:	204641

David

Last edited by factory; 2nd May 2020 at 8:47 pm.
factory is online now  
Old 2nd May 2020, 10:49 pm   #23
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: HP8640B Signal Generator

Check your output amp is OK before spending much.

Those gears look gorgeous!

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 3rd May 2020, 10:50 am   #24
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: HP8640B Signal Generator

My gears look exactly like those David (the plastic ones with splits!).
I still need to remove the attenuator and tuning knobs before I can proceed, but someone has been there before me.
Other gears (eg tuning) are also splitting. I could close the cracks and glue thin washers to the back surfaces but maybe this has to be done in-situ as dismantling might result in more damage.
I suspect the tuning gears have been forced and jumped a tooth or three which might explain the reluctance of the knob to turn more than about 450 degrees. Maybe the two round devices geared to the tuning have reached their end stops?
Today, I'm going to fix the bridge rectifiers. I reckon the original diodes might be equivalent to the 1N4999 3A 200v. Some (but not all) fixing holes are about 1mm dia so I need to work out how to fit 1N5408 diodes (I have a bag of those with 1.3mm dia legs) that should just about allow the board to fit back in place.
If 5 diodes are duff (3 open circuit and 2 with high forward drop) you'd expect other equipments will have the same fault?
If I can see an RF signal it will make further effort worthwhile so cross fingers.
Allan G3PIY
allan is offline  
Old 3rd May 2020, 2:22 pm   #25
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: HP8640B Signal Generator

Progress has taken a step backwards so not too good at present.
I repaired board A12 with new diodes, fitted it and saw all zeroes in the display. Switched off and read up on all zeroes on the display and decided to check the 5.2 volt supply because one red lamp was dimmer than the rest.
Plugged in the kettle lead and was just about to switch on when there was a flash and a loud bang.
The 1.25A fuse was completely black.
The mains switch is in the off position but works normally.
Fitted a new 1A fuse and measured the input with an ohmeter.
Around 10 ohms with 250Kohm to ground.
Checked the voltage setting pcb. This is in the 240v position and is OK.
Looks like the mains filter has shorted to ground leaving a leak.
Mains measured at 249.5 volts.
I hadn't fitted a new C6 yet.
It looks like the mains input assembly is held in place by stainless steel clips?
One set accessible the other hidden by the PSU motherboard.
Allan G3PIY
allan is offline  
Old 3rd May 2020, 2:34 pm   #26
factory
Octode
 
Join Date: Jul 2009
Location: Cheshire, UK.
Posts: 1,706
Default Re: HP8640B Signal Generator

Quote:
Originally Posted by allan View Post
Today, I'm going to fix the bridge rectifiers. I reckon the original diodes might be equivalent to the 1N4999 3A 200v. G3PIY
Have you downloaded any of the service manuals from Keysight?
The A12 rectifier board has different parts listed depending on the age of the unit.

The older 1974 manual lists all 20 diodes as being 1901-0418 (SR1846-12) rated at 400V 1.5A.
But the later 1991 manual gives 16 diodes (CR1 to CR16) as being 1901-0693 (1N4934) rated at 100V 1A & the last four diodes (CR17 to CR20) as being 1901-0662 (MR751) 100V 6A.

David
factory is online now  
Old 3rd May 2020, 2:49 pm   #27
factory
Octode
 
Join Date: Jul 2009
Location: Cheshire, UK.
Posts: 1,706
Default Re: HP8640B Signal Generator

Quote:
Originally Posted by allan View Post
Plugged in the kettle lead and was just about to switch on when there was a flash and a loud bang.
The 1.25A fuse was completely black.
The mains switch is in the off position but works normally.
Fitted a new 1A fuse and measured the input with an ohmeter.
Around 10 ohms with 250Kohm to ground.
Checked the voltage setting pcb. This is in the 240v position and is OK.
Looks like the mains filter has shorted to ground leaving a leak.
Mains measured at 249.5 volts.
I hadn't fitted a new C6 yet.
It looks like the mains input assembly is held in place by stainless steel clips?
One set accessible the other hidden by the PSU motherboard.
Allan G3PIY
Do you have a picture of the mains filter? the type fitted may vary with the age of the unit. Could it be a Schaffner line filter it's a newer unit? these are known to fail badly, also have you disconnected the RIFA capacitor (C6?)?

David
factory is online now  
Old 3rd May 2020, 4:07 pm   #28
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: HP8640B Signal Generator

I managed with a lot of levering to detach the filter/mains setting panel. The leads are very short and the whole thing is enclosed in a pop riveted metal cover.
The easiest solution is to fit a new panel with an IEC connector (maybe with a built-in filter if I can find one in the junk box) and fuse and to hardwire the transformer tappings, back at the tags on the transformer, for 240 volts.

My A12 board has 20 identical diodes which I think are 1N4999.
Allan
Attached Thumbnails
Click image for larger version

Name:	filter.jpg
Views:	131
Size:	181.1 KB
ID:	204725   Click image for larger version

Name:	A12bef.jpg
Views:	133
Size:	67.7 KB
ID:	204727  
allan is offline  
Old 3rd May 2020, 4:28 pm   #29
factory
Octode
 
Join Date: Jul 2009
Location: Cheshire, UK.
Posts: 1,706
Default Re: HP8640B Signal Generator

I've got one of those line connectors in the spares box (part no #0960-0443), never realised it was a line filter too, most of the connections unplug apart from the three mains wires.
Click image for larger version

Name:	SAM_3319a.jpg
Views:	121
Size:	76.5 KB
ID:	204729

What is the HP part number on those diodes? neither of the two manuals I've looked at have 1N4999 as the part, but it is in the same package style as those, just higher rated (except the four 6A diodes used in the 1991 version, see previous post).

David
factory is online now  
Old 3rd May 2020, 5:10 pm   #30
factory
Octode
 
Join Date: Jul 2009
Location: Cheshire, UK.
Posts: 1,706
Default Re: HP8640B Signal Generator

There appears to be at least three versions of the A12 rectifier board;

SN prefix 1229A to 1249A uses 08640-60003, this is the oldest & has a crowbar circuit across the first bridge rectifier.

Yours uses 08640-60326, no idea which prefix range they cover as two free PDF manuals don't mention this part, the crowbar components are not present but the parts are still labelled on your board.

SN prefix 2520A to 2923A uses 08640-60375, this is shown in the 1991 manual, it does away with the can style diodes for more modern ones, the 1991 manual also has a mistake as it shows 08640-60326 on the circuit diagram.

With most HP stuff you really need the correct manual for the SN prefix of your unit, it can sometimes cause problems & confusion if you don't.

David
factory is online now  
Old 3rd May 2020, 5:20 pm   #31
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: HP8640B Signal Generator

I decided I'd nothing to lose removing the pop rivets David.
Inside the case is a small ferrite ring wound with a few turns for live and neutral wired to the two feed-through capacitors (those brown things sticking out the case). One of these had broken down. I put the C6 failure and the filter failure to the use of 250 volt working components. My mains supply can be well over 250 so the plan now is to solder wires direct to the mains socket connections, add a couple of decoupling capacitors, and run a pair of wires through the holes left by the feed throughs to an external mains filter using better filter capacitors.
Self tapping screws in place of the pop rivets. It saves messing with the transformer wiring and leaves the voltage selector intact.
cheers
Allan
allan is offline  
Old 3rd May 2020, 5:29 pm   #32
factory
Octode
 
Join Date: Jul 2009
Location: Cheshire, UK.
Posts: 1,706
Default Re: HP8640B Signal Generator

That explains why it went bang then.

The RIFA capacitors don't fail for that reason, they fail because they are metallised paper in an epoxy case, the two materials expand at different rates, ageing then causes the casing to crack and moisture can get to the paper part, a few decades later they turn into a horrible stink-bomb.
A forum search for RIFA Y, X or X2 capacitor should find lots of threads about these crap capacitors. Always check for them before powering up old equipment.

In my 8640B, a previous owner let the smoke out of this one;
Click image for larger version

Name:	SAM_0646a.jpg
Views:	129
Size:	75.1 KB
ID:	204748

David
factory is online now  
Old 3rd May 2020, 5:59 pm   #33
Alan_G3XAQ
Pentode
 
Join Date: Apr 2018
Location: Canterbury, Kent, UK.
Posts: 189
Default Re: HP8640B Signal Generator

Quote:
Originally Posted by allan View Post
saw all zeroes in the display. Switched off and read up on all zeroes on the display and decided to check the 5.2 volt supply because one red lamp was dimmer than the rest.
Allan G3PIY
I've seen this on one of my 8640Bs and it was just the TO3 pass transistor that is on the rear apron of the chassis had dirty leads. Yes, the transistor is a plug in, not a solder in. After a bit of a clean with IPA all became bright and beautiful again.

I'm sorry you are having a torrid time of it getting the old girl up and running. I got some replacement brass gears from a guy in the USA who regularly posted on the HP forum when it was on Yahoo. That was quite a while ago. They moved the forum to groups.io but I've no idea if the chap is still able to offer new gears.

https://groups.io/g/HP-Agilent-Keysi...pment/messages

73, Alan G3XAQ
Alan_G3XAQ is offline  
Old 3rd May 2020, 6:42 pm   #34
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: HP8640B Signal Generator

Favourite thing with a disappointing display on an 8640 was the external 10MHz ref switch being knocked into the ext position when someone carrying the thing slipped a hand round the back for extra support. Amazing what hyper-intelligent pan-dimensional beings got caught out by that one

Most 8640 electronic problems are in the power supplies.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 4th May 2020, 10:43 am   #35
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: HP8640B Signal Generator

I managed to replace the filter feed-through capacitors with ceramic feed throughs and fit a pair of 400VAC filter capacitors external to the filter and refitted the assembly. All told a really horrible job.
Removed the plug-in 2N3055 and cleaned tarnish/corrosion from its legs and refitted it.
Switched on and saw all zeroes (again). Checked the voltages which were all respectable. Pressed various buttons and the bottom left of the group of six (without a label) resulted in a numeric display. As the top section of the panel wasn't fitted I hadn't spotted it said INT/EXT.
The limited rotation of the tuning knob gives me a reading 3352 to 4168.
Almost as if the last owner had been using it on the 80m band before it expired?? The knobs except tuning and attenuation are detached so its not easy to check the range switch setting and that shaft is very stiff so I refrained from forcing it.
The meter reads zero but turning off the 8640 makes the needle jump so the movement is OK.
Progress!
Allan
allan is offline  
Old 4th May 2020, 12:20 pm   #36
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: HP8640B Signal Generator

Ah, that's the int/ext input for the counter to count. On the back panel is a little toggle that switches the counter gate from the internal crystal reference to an external one. Flipping this is more subtle to find than the one on the front panel. But that's good it's counting.

Proves the cavity oscillator and dividers are going.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 4th May 2020, 1:20 pm   #37
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: HP8640B Signal Generator

A few faults to sort out.
I'm a bit concerned about the zero meter reading in the RF position and very little reading in FM settings.
The AM mod switch is seized with the lever not engaged but that probably isn't important yet..
If I turn the range switch the frequency reading doesn't change it stays at 04168x with x variable with fine tune.

When the tuning knob is turned clockwise it stops dead and you can see all the plastic tabs lined up. fuzzy picture-sorry
As the knob turns anticlock the back edge of the first tab blocks further movement after something a little more than one revolution. slightly better picture...Is it just a case of it being forced in the past? Frequency changes from 4168 to 3352 (or vice versa... I can't remember)
Allan
Attached Thumbnails
Click image for larger version

Name:	end.jpg
Views:	93
Size:	28.3 KB
ID:	204860   Click image for larger version

Name:	end2.jpg
Views:	81
Size:	35.8 KB
ID:	204861  
allan is offline  
Old 4th May 2020, 1:56 pm   #38
Nymrod121
Nonode
 
Nymrod121's Avatar
 
Join Date: Apr 2012
Location: Worcestershire, UK.
Posts: 2,039
Default Re: HP8640B Signal Generator

Just to mention that on mine, the output level is only indicated when the 'RF output' switch is operated (adjacent to the output N-type).
__________________
"What a depressingly stupid machine." [Marvin: HHGTTG]
Nymrod121 is offline  
Old 4th May 2020, 2:31 pm   #39
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: HP8640B Signal Generator

That switch is very rough but it does actually work and still no reading on the meter.
I'll see if I can trace where the wires go from the switch

I thought I'd check the RF output and the thing delivers a nice sinewave at 4MHz at about 5volt RMS and the attenuator is OK, so the RF output must be working...
Allan

Last edited by allan; 4th May 2020 at 2:49 pm.
allan is offline  
Old 4th May 2020, 3:10 pm   #40
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: HP8640B Signal Generator

Working RF output is brilliant.

Does the frequency coming out tune with the main tuning control, and does it change when you move the range switch?

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 7:33 pm.


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 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.