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 21st Jun 2020, 9:26 pm   #41
Alan_G3XAQ
Pentode
 
Join Date: Apr 2018
Location: Canterbury, Kent, UK.
Posts: 189
Default Re: Replacement for HP8640B signal generator

Yes, Jeremy. I've PM's Woodchips. I'd like the sig gen but the spectrum analysers aren't quite what I want. Instead I'd prefer the TK2373 range extender for my TF2730 if I can get one from Stewarts. Or should I go for the HP8568A and put my 2730 on eBay?

Suddenly I am overwhelmed with options. Like buses.
Alan_G3XAQ is offline  
Old 21st Jun 2020, 9:37 pm   #42
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Replacement for HP8640B signal generator

I've not used the older A version of the HP8568 analyser but I think the RF performance is the same as the B. I think it is just the processing that is slower. The Mi 2370 is a classic instrument and very capable in the right hands. However, the HP8568 is so good it is hard to believe it was released in 1978. It's as if HP had a time machine back then. To have been on the design team for that instrument must have been amazing because it was so ahead of its time in so many ways.

If you can live with the size and power and the fan noise then it is going to be light years ahead of the TF2370 with the extender. A completely different experience.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 21st Jun 2020, 9:44 pm   #43
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Replacement for HP8640B signal generator

I do think you should try and maintain at least one HP8640B sig gen because I think that sig gen was designed to meet some very strict (spectral purity) requirements that make it fairly unique.

A healthy 2019 is better than a lot of modern stuff when it comes to close in phase noise across about 2.5 -30MHz. However, as I said in an early post it won't withstand the same level of scrutiny as the 8640B. In other words, if you want to look for (curious?) noise bumps and unwanted spurious near the noise floor then the 2019 won't be as well behaved as the 8640. With a HP8568 spectrum analyser on hand to compare the two sig gens you will find that out quite quickly.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 21st Jun 2020, 9:58 pm   #44
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Replacement for HP8640B signal generator

Forgive me for tempting you further towards the HP8568A, but maybe you should read the HP Journal from June 1978

https://www.hpl.hp.com/hpjournal/pdf...Fs/1978-06.pdf

Scroll down the page in the link below to get an idea of the impact this instrument caused when it was first displayed in 1978.

http://hpmemoryproject.org/wb_pages/wall_b_page_04.htm

I've got an HP8568B here and it is the ex works analyser that was donated to the company about 30 years ago. It has had a hard life and was used all day every working day for over 15 years. I think I bought it from the company in about 2005. The CRT went fuzzy and a bit vague but I managed to fix this issue a few years ago. If the CRT fails then this can be quite a problem because a replacement CRT is going to be hard to find. But for the cheap deal you are being offered why even worry about this?
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 21st Jun 2020, 10:04 pm   #45
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Replacement for HP8640B signal generator

Note that the 8566 and 8568 share the same top box (display, later IFs etc) and that it is the bottom box which is different.

As always, with HP instruments from before the middle 90s, have a look in HPJ for some insight. Th 8568 LO scheme is rather special. Essentially it has a parallel strip running backwards as a sort of tracking generator would, and it is this which is phase locked.

The 8566 is more normal. Well, it set the norm.

There are a few small differences like a different crystal osc freq standard, but it's really a matter of substantially more processing power leading to more IO capability and some extras in the display.

On the whole, the B is worth having if one is available. It will also be younger.

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 21st Jun 2020, 11:33 pm   #46
factory
Octode
 
Join Date: Jul 2009
Location: Cheshire, UK.
Posts: 1,706
Default Re: Replacement for HP8640B signal generator

Quote:
Originally Posted by G0HZU_JMR View Post
I've got an HP8568B here and it is the ex works analyser that was donated to the company about 30 years ago. It has had a hard life and was used all day every working day for over 15 years. I think I bought it from the company in about 2005. The CRT went fuzzy and a bit vague but I managed to fix this issue a few years ago. If the CRT fails then this can be quite a problem because a replacement CRT is going to be hard to find. But for the cheap deal you are being offered why even worry about this?
An LCD conversion kit has been available for a few years, it's not cheap though, no idea how good it is either.

David
factory is online now  
Old 22nd Jun 2020, 1:00 am   #47
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Replacement for HP8640B signal generator

I've wondered about that kit.

The existing display is created digitally, then realised by vector generators and an XYZ analogue display.

Does the kit just digitise the vector generator analogue signals, or does it intercept the digital info upstream of the DACs?

There is a fast display mode where analogue trace info is routed directly for very fast scan times. I suspect the LCD doesn't attempt this.

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 22nd Jun 2020, 9:32 am   #48
Alan_G3XAQ
Pentode
 
Join Date: Apr 2018
Location: Canterbury, Kent, UK.
Posts: 189
Default Re: Replacement for HP8640B signal generator

A dumb non-owner question. From reading the HPJ article it's not obvious that the HP8568A comes with a tracking generator. It's a must-have when looking at filters and suchlike.

Alan
Alan_G3XAQ is offline  
Old 22nd Jun 2020, 10:49 am   #49
woodchips
Octode
 
Join Date: May 2010
Location: Grantham, Lincolnshire, UK.
Posts: 1,172
Default Re: Replacement for HP8640B signal generator

No tracking generator in the 8568A.

The 8568A has BNC connectors on the back for the CRT display. HP made the 143A scope with a 10" screen and 20MHz response, did have one. I have a Kikusi large screen display you can have but don't think it has the bandwidth, need 5MHz to display the 8568A, possibly a Tek 7603 with amplifier rather than timebase might work, get the large screen as well.

The 8568A doesn't work, was busy fixing display PSU problems and ran out of strength to keep turning it over. Got some extender cards for it somewhere.
woodchips is offline  
Old 22nd Jun 2020, 11:11 am   #50
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Replacement for HP8640B signal generator

It is possible to use an older (external) HP tracking generator with the 8568 as in this video.However, it requires an extra LO at 500MHz as shown in the video.

https://www.youtube.com/watch?v=_yZOW_aABlk


I'm not that keen on internal or external tracking generators having used lots of them with various spectrum analyser models. For filter designs up to 50MHz and maybe even 150MHz there is always the little nanovna. I think the larger version with 4.3" screen costs about £100. This can manage about 80dB range across the HF bands and it can also measure impedance.

However, if the 8568A is faulty then I guess that may change things. The 8568 and the 2019 are both quite old instruments and will probably require some fairly regular maintenance.

Having owned a couple of 2019s since the mid 1990s I can strongly suggest that some of the unreliability is due to thermal issues. These generators don't have a fan and they can get quite hot inside unless used with good ventilation. Also a lot of them were rack mounted and they lose their standoff feet on the lower face of the chassis. This means they block their own vents when sat on a flat surface. I found that once I started allowing the generators to ventilate properly they became much more reliable. I don't think either of my 2019s has developed a fault since. I'm guessing that they have been reliable for about 15 years now although I may have forgotten. Most of the faults with mine were intermittent connections or failed electrolytic caps. When the caps fail the close to carrier spurious levels start to creep up. Other areas of weakness are the internal 10MHz OCXO as it can fail (repairable!) and also the VCOs in the VCO die cast box can fail. Usually this is as simple as a failed capacitor.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 22nd Jun 2020, 11:35 am   #51
Alan_G3XAQ
Pentode
 
Join Date: Apr 2018
Location: Canterbury, Kent, UK.
Posts: 189
Default Re: Replacement for HP8640B signal generator

I'm picking up the 2019A from Woodchips (Bob) next week.

Someone else has offered me a fully working example of the later 8568B model and with it the 8444A tracking generator. It's twice what a TK2373 extender would cost me but is still only 1% the original purchase price. I suppose I can recover some of the outlay by selling off the old stuff.

I hope my workbench is strong enough for all this stuff.

73, Alan G3XAQ
Alan_G3XAQ is offline  
Old 22nd Jun 2020, 11:50 am   #52
Alan_G3XAQ
Pentode
 
Join Date: Apr 2018
Location: Canterbury, Kent, UK.
Posts: 189
Default Re: Replacement for HP8640B signal generator

Jeremy,

AFAIK the 8444A being offered is the one with the 500MHz LO to match the 8568B.

Yes, I have a small-screen nanoVNA but it isn't a bench instrument and has somewhat flaky PC viewer software as you have previously noted. I confess I find my 2370 and return loss bridge more convenient for both transmission and reflection measurements on LC filters. The frequency resolution on the Nano is good for measuring motional parameters of crystals and ceramic resonators, and composite filters made from same, and is great for aerial measurements in the field.

73, Alan

[I would put off-hand comments like this in a PM to avoid cluttering hte more general discussion but you seem to have that feature disabled]
Alan_G3XAQ is offline  
Old 22nd Jun 2020, 12:00 pm   #53
Alan_G3XAQ
Pentode
 
Join Date: Apr 2018
Location: Canterbury, Kent, UK.
Posts: 189
Default Re: Replacement for HP8640B signal generator

Quote:
Originally Posted by G0HZU_JMR View Post
as shown in the video.
It looks very nice. But yes, I can hear the Boeing 747 jet engine fans!

73, Alan
Alan_G3XAQ is offline  
Old 22nd Jun 2020, 12:16 pm   #54
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Replacement for HP8640B signal generator

As David says the B version is much nicer and faster. Probably the first thing to learn with the HP8568B is how to do the RECALL 8 and RECALL 9 adjustments and also the internal (automated) calibration optimisation/correction routine. This is launched with (blue) [SHIFT] then [FREQUENCY SPAN] and it takes about a minute to self correct across various RBW filters and attenuation settings.

Each time you power cycle the machine you can recall the previous calibration correction by pressing [SHIFT] then [START FREQ] and a little Corr'd logo will appear on the CRT display. It's worth learning these things if you want to get the very best out of an already brilliant analyser.

Quote:
It looks very nice. But yes, I can hear the Boeing 747 jet engine fans!
I also have the 22GHz HP8566B here and the fan on the back of the RF section is huge. I have never done a side by side comparison but I think the 22GHz 8566B is even noisier! For general RF work up to 1500MHz I would always choose the 8568B even though the 22GHz 8566B is often described as the high performance version. The dynamic range and the phase noise is much better on the 8568B.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 22nd Jun 2020, 3:06 pm   #55
Alan_G3XAQ
Pentode
 
Join Date: Apr 2018
Location: Canterbury, Kent, UK.
Posts: 189
Default Re: Replacement for HP8640B signal generator

Hmmm, the 8444A I'm being offered has a dud output amplifier/ALC module so the owner is taking a TG signal from the input to that module. AIUI it is LSI (like the HP8640B) and replacements are unobtainable other than from another donor. I expect the TG signal is 20dB down on expectations and is not constant amplitude over the full frequency range, although it is claimed to be "OK" over a narrow range.

Ah well, you said you don't like TGs anyway

Alan
Alan_G3XAQ is offline  
Old 22nd Jun 2020, 3:51 pm   #56
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Replacement for HP8640B signal generator

Quote:
Originally Posted by Alan_G3XAQ View Post
A dumb non-owner question. From reading the HPJ article it's not obvious that the HP8568A comes with a tracking generator. It's a must-have when looking at filters and suchlike.
THe 8568 does have what amounts to a tracking generator inside it. It's a major part of how the synthesisers work. It is a major difference to what's in the 8566 and all the other analysers.

However, it is NOT brought out for the user, and it's not easily modified to be so.

So users wind up having to buy an external accessory box as a tracker.

Sort of you can't have your cake and not eat it!

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 22nd Jun 2020, 3:58 pm   #57
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Replacement for HP8640B signal generator

Many network analysers are not too good on phase noise, while it's a prime parameter on spectrum analysers.

As a consequence, very narrow filters can resolve the noise sidebands on a network analyser, and the resultant plots show undergrowth at the lower part of the transition region that aren't really there on the filter. So most VNAs come a bit of a cropper plotting crystal filters. Spectrum analysers are often better here. One of the good guys on the VNA front is the 3577A, but that's only usable to 200MHz.

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 22nd Jun 2020, 6:32 pm   #58
Alan_G3XAQ
Pentode
 
Join Date: Apr 2018
Location: Canterbury, Kent, UK.
Posts: 189
Default Re: Replacement for HP8640B signal generator

Seeing as the HP8445A works by "just" mixing, am I right in thinking the overall SA+TG phase noise performance will be governed by the SA? Well, mostly.

The nanoVNA uses a Si5351A oscillator/cock chip, which is around 15dB noisier close-in than the Si570, which itself isn't stellar. OTOH if they used something fancy like an AD9912 the price of the bare chip would be almost as much as the whole nanoVNA. You pays your money I guess.

73, Alan
Alan_G3XAQ is offline  
Old 22nd Jun 2020, 6:50 pm   #59
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Replacement for HP8640B signal generator

Yes, to all of the above!

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 22nd Jun 2020, 10:49 pm   #60
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Replacement for HP8640B signal generator

If it helps the typical system phase noise of the HP8568B will look like the hastily drawn blue trace below when the analyser is used on spans of 1MHz or less and at frequencies below about 200MHz. The phase noise will look like a pedestal of noise about 140kHz wide at -112dBc/Hz. This is very clean for a >40 year old design. However, this is the overall (corrected) system phase noise for the whole LO system on spans of 1MHz or below. This does not mean the LO1 phase noise will be this clean because the LO1 signal is only part of a fairly complicated synthesiser.

I've never used the 8444 tracking generator but I think it has a free running cavity oscillator inside it at about 1.5GHz. Therefore, I would guess that the phase noise of the Tgen source will contain the close in noise of the free running cavity oscillator plus whatever phase noise is riding on the LO1 signal. This LO1 signal is then fed to the tracking generator. I think it will still be quite good but I don't think it will be as clean as the image below at offsets below a few kHz.

Has anyone tried to look at the phase noise of this tracking generator on a narrow span when used with an HP8568B? I can only guess at what it might be.
Attached Thumbnails
Click image for larger version

Name:	HP 8568B PN.gif
Views:	63
Size:	5.1 KB
ID:	209280  
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 12:23 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 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.