The HP 5246L counter

ralph klimek 26-may-2009

this page created to pay tribute to a classic piece of Hewlett Packard test and measurement equipment

copyleft: the text and images, are mine, and I say you are free to do with them as you wish with the exception of commercial trademarks.

This item was found in a heap of hard rubbish that had been discarded by the chemistry department at Monash. When I chanced upon this junk heap  it had been raining solidly for about a week and this junk heap had got quite wet.  Underneath some real rubbish I found this HP frequency counter.  I do not need yet another counter, I allready have a very nice homemade unit and  a nice General Radio 7 digit counter with Nixie Tubes.   Here was a counter, a large rack mount device, but it had large Nixie tubes  and nothing looks more official than Nixie Tubes. I took it, really for the display tubes.  (I just want my ham shack to look like the Centre 0' Global Domination too)

The office here is the usual dehumidified cube farm, and in one week, the dry atmosphere here had removed all traces of water from the insides and at that point I took it home.  I had to improvise a power connector, this old style of removeable power cord cannot now be found.  I powered it on, not really expecting anything usefull to happen, but I powered it on , at least so I could reassure myself that I had gone to at  least  this small amount of trouble.  

It worked !  Not only did it work,  it worked perfectly.   This equipment, from what I have been able to find out was made about 1963.   A more advanced  companion unit is described in the Hewlett-Packard Journal and has a price list as well.  The list price corresponded to what would have been twice the cost of a suburban Melbourne house in 1963 and as such I stood as much chance of owning  one as I had of becoming the King.  It was the sort of thing that major governments would purchase for weapons research and other such worthy causes.  

There you have it, biult in 1963 (I was in kinder, watching grainy newsreals of Project Mercury with this thing in the background!) it had seen years of service at Monash, and been left exposed to  public abuse and the rain....and it just worked.  Thats nearly 50 years.

There were no integrated circuits back at the time this equipment would have been designed.  It is all discrete silicon and germanium transistors.  It can count natively to 50Mhz.  It can count to 3Ghz with the appropriate plug-in, which by good fortune was also found underneath some banana peels and ash tray contents in the dumpster.  The BCD counters drive an most amazing BCD to decimal nixies tube driver that consists of neon bulbs over a special array of cadmium sulphide photoresistive cells that perform the logical operation of converting a BCD pattern of Neons into a UNARY signal for driving the appropriate anodes in the Nixie tube.

There is one and only one integrated circuit in this unit. It is on the input preamplifier board and I can only conclude that it is a differential amplier, probably on of the very first ones.  The VHF/UHF prescaler plugins work on the principal of downconversion to something below 50Mhz which can be counted.  The internal 10Mhz timebase is fed to a step-recovery diode which generates a good harmonic comb. A tuneable cavity selects the required harmonic which is mixed with the input signal, the differance being counted.  A step recovery diode cost only $USD115 back then, a couple of months average wages.

The unit can supply data to a computer and be controlled by one through a parrallel BCD data bus which is brought out to the back. This must have seemed ridiculously far-sighted back then.

It must have seemed like alien technology back then, considering the state of the art in commercial and consumer electronic goods.

Here are some pictures. No complete manual seems to be available , allthough, manuals for the plugins are widely available.(see BAMA) The service manual that appears by the very generous courtesy of Agilent is ,sadly , not complete and is only the user manual. It only contains an internal block diagram.

HP 5246L counter HP 5246L counter assorted plugins
the black boxes are the Neon CdS BCD-unary  Nixie decoder/drivers This discrete logic works at 50Mhz ! It took till 1980 before 74x TTL logic could! a business-like but noisy fan
oven ensures excellent timebase stability bottom view of card bus there is one and only IC in this equipment.
Its on this preamplifier PCB, prabably a diff amp
only one slightly burned resistor in the oven temperature regulator the 50 way centronics connecter permits connection to a 1963!
regulators I dont know what the relay does behind the plugin sub chassis
internal test, I count,
 I work !
clicky on image above. this is an HD  MP4 movie of nixies tubes in action.
warning.. file is BIG 80Mbytes and not very interesting but see it if you really want to see NIXIE tubes in action, taken with a Cisco Flip camera I won in a raffle!

scanned manuals are now appearing on the web.  Agilent has the user manual for this beastie, a very old and battered one at that,
but it does give a good description of how it works. Better still  near equivalent is available on this most remarkable HP Museum site.
It is not the same equipment but is sufficiently similar to permit fixing most faults.

Check out this most-cool site, a huge amount of work has gone into this HP tribute site.

Sadly, the oven heater has completely failed. The oven was designed to be non repairable and has been potted with some horrible expanding foam.  There is a mention of it on the eevblog. Never mind. Repair, replace or restore ?  The frequency reading is now off by a couple kilohertz.  Perhaps I should just place a modern equivalent in the same housing ?

More to come.
I am also the proud custodian of a HP model 140a  Oscilloscope which still works "perfectly"; allthough some pots and wafer switches  are a bit noisy now. Its got the standard timebase plugin and the standard 5Mhz dual channel amplifier plugin. The huge screen is still brilliant. It is a  thermionic tube, transistor hybrid circuit, which like all the classical HP gear was simply 10 years ahead of its time. When I have some time I will post what documentation I have gathered and images.

created Mon May 25 17:49:14 EST 2009
misc spelling,  added external site reference Fri Oct  1 17:50:25 EST 2010;
added nixies tube movie Wed Jan  5 13:20:47 EST 2011