Author Topic: HP 5316A Teardown and Frequency Standard  (Read 57300 times)

0 Members and 2 Guests are viewing this topic.

Offline amc184Topic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: nz
HP 5316A Teardown and Frequency Standard
« on: October 17, 2014, 05:14:31 am »
I bought this HP 5316A from eBay, and it was another classic eBay purchase.  It was sold as powering on but otherwise untested.  When I got it, it partially worked but threw fault codes and was really, really sticky and brown.  Maybe it was used by a smoker?  It cleaned up nicely and the internals were pristine, so for USD36 it was a great deal.





I replaced the electrolytics and tantalums and the counter worked correctly.  The only other thing that bothered me about it was that it was just the base model with a basic crystal as the frequency standard.  There were two frequency standard options for the 5316A; 001 (a TCXO) and 004 (an OCXO).  Looking at the service manual, these options were pretty simple, mainly just a regulator to bring the 8 to 10VDC standby supply down to whatever the TCXO or OCXO needs (5VDC in both cases).  I designed a small PCB with a regulator and a Morion MV85A OCXO.  These are available cheaply on eBay and require a 5VDC supply.  Most require a 12VDC supply, and I really didn't want to have to have a step up supply, especially as the oven draws a fair bit of current on startup.







The PCB works well, it mounts on standoffs right where the original oscillator card sits.  The calibration pot lines up okay with the hole in the rear panel, now I just have to figure out how I'm going to calibrate it ....

I have some spare PCBs as well, if you have a 5316A and want one PM me.  I think it should work in a 5316B as well.
 
The following users thanked this post: kj7e, dtryon

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: HP 5316A Teardown and Frequency Standard
« Reply #1 on: October 17, 2014, 06:03:50 am »
These are fabulous counters.  I have an HP 5315A with the TCXO option which can be seen edge on in the photo below.

I calibrated mine using the 5 Hz timing output on my GPS and a long gate time.
 

Offline psykok

  • Frequent Contributor
  • **
  • Posts: 289
  • Country: fr
Re: HP 5316A Teardown and Frequency Standard
« Reply #2 on: October 17, 2014, 06:40:08 am »
Hi,

Great work, I have the same counter but with TCXO option.


Alex
 

Offline daqq

  • Super Contributor
  • ***
  • Posts: 2321
  • Country: sk
    • My site
Re: HP 5316A Teardown and Frequency Standard
« Reply #3 on: October 17, 2014, 07:18:40 am »
Good job! Thanks for the peek inside!
Believe it or not, pointy haired people do exist!
+++Divide By Cucumber Error. Please Reinstall Universe And Reboot +++
 

Offline nixxon

  • Frequent Contributor
  • **
  • Posts: 342
  • Country: no
Re: HP 5316A Teardown and Frequency Standard
« Reply #4 on: November 11, 2014, 11:24:53 pm »
I have agreed with a dude to pick up a HP 5315A with option 003 (1 GHz Channel C).

I am thinking about retrofitting a OCXO. I have noticed that (i.e. reading the last page here: https://www.valuetronics.com/Manuals/HP_5315A_HP_5316A.pdf) that the option 004 "High Stability OVEN Time Base" differs between the 5315A and the 5315B/5316A.

Do you think that the DIY option 004 equivalent that fits the 5315B/5316A will fit the 5315A anyway?
 

Offline amc184Topic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: nz
Re: HP 5316A Teardown and Frequency Standard
« Reply #5 on: November 12, 2014, 06:16:59 am »
It looks to me like the form factor of the plug in frequency standard modules is the same for both the 5315 and 5316.  The problem with using the PCB I made in the 5316 is the black plastic pillar to the left of the two holes the PCB mounts to.  This will interfere with the top left corner of the PCB, where 'REV B1' is written.  You could just file that corner off, though you might have to route the RF output trace if you cut that far into the PCB.
 

Offline nixxon

  • Frequent Contributor
  • **
  • Posts: 342
  • Country: no
Re: HP 5316A Teardown and Frequency Standard
« Reply #6 on: November 12, 2014, 09:21:23 am »
Maybe option 004 for the plastic cased 5315A is less power hungry than option 004 for the other two stackable/rackable, metal chassis models (5315B and 5316A), because the 5315A may be run on batteries. For some reason, option 004 for the 5315A originally cost $460, whereas option 004 for the stackable/rackable models cost significantly more, $605.

amc184: Is the size and shape of your DIY option 004 PCB identical to that of the standard A7 assembly that was replaced in the 5316A?

(in the 5315A, option 004 replaces the standard A13 oscillator assembly)
« Last Edit: November 14, 2014, 09:42:16 am by nixxon »
 

Offline nixxon

  • Frequent Contributor
  • **
  • Posts: 342
  • Country: no
Re: HP 5316A Teardown and Frequency Standard
« Reply #7 on: November 13, 2014, 02:43:51 pm »
It looks to me like the form factor of the plug in frequency standard modules is the same for both the 5315 and 5316.  The problem with using the PCB I made in the 5316 is the black plastic pillar to the left of the two holes the PCB mounts to.  This will interfere with the top left corner of the PCB, where 'REV B1' is written.  You could just file that corner off, though you might have to route the RF output trace if you cut that far into the PCB.

I just picked up my 5315A. It looks very nice and clean on the inside. The standard oscillator circuit board (assembly) is standing perpendicular to the main board on it's long edge. The oscillator circuit board  is ~75 mm long and ~31 mm tall. There are 6 pins connected to the main board. To access the frequency adjust screws, the instrument top cover must be removed.

Do you think the "HP 5316A OPTION 004 EQUIVALENT" board will work with the 5315A as well? (With some fastening mods).
 

Offline amc184Topic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: nz
Re: HP 5316A Teardown and Frequency Standard
« Reply #8 on: November 13, 2014, 10:22:57 pm »
The oscillator boards in the 5315 and 5316 are the same form factor, and look to be the same circuit.  The component layout is a bit different, allowing for closed case calibration in the 5316.  However, the replacement module I've made is not quite the same as the original board.  While the original PCB sits vertically, mine sits horizontally.  I dismantled my 5316A to get some more shots, this time with the side rail removed for clarity.

Original clock module:



My replacement OCXO module:



 The component side of the original 5316A module:



The main reason I chose to change the orientation like that was because I have no idea where to get those little rivets o angle brackets on the original module.  Even if I could, they're not really suitable for the greater weight of the OCXO; the original option 004 PCB only held the voltage regulator, the OCXO was mounted separately on a bracket.  My module mounts onto two standoffs, which screw into the original threaded holes in the PCB.

So to amble back around to your question, my answer from yesterday hasn't changed.  If you want to use this board, you'll need to modify it to accommodate the large plastic pillar in the 5315 (the one underneath the text '~31mm' in the second picture of your last post).
 

Offline amc184Topic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: nz
Re: HP 5316A Teardown and Frequency Standard
« Reply #9 on: November 13, 2014, 11:00:50 pm »
I've had a look at the HP 5315A/B service manual, and it looks like the 6 pin header's pinout might be different to a 5316A/B.  However, that same manual also suggests that a 5315 would have the basic oscillator built into the mainboard, and a plug in module would only be used for the options, so it clearly doesn't match your 5315A.  Here's the pinout for the 5316A and my module:

1) VCC (roughly 10VDC, unregulated)
2) RF
3) GND
4) NC
5) NC
6) VCC (roughly 10VDC, unregulated)

It'd be good if you could have a look at your 5315A and determine its pinout.
 

Offline nixxon

  • Frequent Contributor
  • **
  • Posts: 342
  • Country: no
Re: HP 5316A Teardown and Frequency Standard
« Reply #10 on: November 13, 2014, 11:18:46 pm »
Thank you for the information and pictures. Very interesting. There are a few different versions of the 5315A/B manual, corresponding to different serial prefixes. The manual covering my unit has the 2120A serial prefix and is printed in 1981. I will figure out the pins in my unit, to compare with the 5316 pinout.

For some reason, the 5315A/B manuals I have found are missing the service section (VIII).
The 5316A manual does however include Section VIII with Service info and the schematics of the OEM OCXO as well.
« Last Edit: November 14, 2014, 11:32:09 am by nixxon »
 

Offline nixxon

  • Frequent Contributor
  • **
  • Posts: 342
  • Country: no
Re: HP 5316A Teardown and Frequency Standard
« Reply #11 on: November 14, 2014, 08:03:08 pm »
I've had a look at the HP 5315A/B service manual, and it looks like the 6 pin header's pinout might be different to a 5316A/B.  However, that same manual also suggests that a 5315 would have the basic oscillator built into the mainboard, and a plug in module would only be used for the options, so it clearly doesn't match your 5315A.  Here's the pinout for the 5316A and my module:

1) VCC (roughly 10VDC, unregulated)
2) RF
3) GND
4) NC
5) NC
6) VCC (roughly 10VDC, unregulated)

It'd be good if you could have a look at your 5315A and determine its pinout.

The pinout of the 5315A is the same as the 5316A (pin 1 closest to the corner):

1) ~+8.8 V (with 600 mV sawtooth ripple)
2) 10 MHz (Fluke 87: ~+1.44V)
3) GND
4) +5.00 V
5) NC (+0.001V)
6) ~+ 9.5V

Enclosed is a screenshot of the signal on pin 2. My Rigol 1102 (1052 with mod) oscillator is in sync with the 5315A oscillator, showing 10.0000 MHz and sometimes 9.99999 MHz). The Rigol Vrms Offset is identical to that of the Fluke 87: +1.44V.

I now wish I had an "HP 5316A OPTION 004 EQUIVALENT" PCB to populate with, along other components, my recent eBay order of a "MORION MV85A OCXO"...
« Last Edit: November 14, 2014, 11:16:21 pm by nixxon »
 

Offline amc184Topic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: nz
Re: HP 5316A Teardown and Frequency Standard
« Reply #12 on: November 14, 2014, 09:10:43 pm »
It looks like the PCB will work.  The pinout for the earlier 5315 was:

1) NC
2) RF
3) GND
4) VCC (5V regulated)
5) NC
6) NC

So it looks like your later 5315 is designed to allow the use of either type of module.  I'll PM you about getting a PCB shipped to you.  I've also attached the schematic, design files and gerber.  Anyone should feel free to use them to make their own PCBs, or modify them to suit their own counter.  They're made with DipTrace, the type numbers in the bill of materials are element14 part numbers.
 
The following users thanked this post: beastor

Offline nixxon

  • Frequent Contributor
  • **
  • Posts: 342
  • Country: no
Re: HP 5316A Teardown and Frequency Standard
« Reply #13 on: November 14, 2014, 10:00:57 pm »
That is very kind of you. Thank you very much  :-+

It seems that my Siglent SDG1020 is the one instrument of mine that is out of sync (18 ppm). In that regard, I am looking forward to receiving a $25 "1PCS Vanguard TCXO 0.1ppm 25.000MHz Ultra precision Golden Oscillator AQ" that I ordered the other day from along1986090 on eBay. Both this and the other oscillator (OCXO for the HP) hopefully arrive in time for the Christmas holidays.

Some good reading on "Understanding Frequency Counter Specifications" by HP:
http://www.leapsecond.com/pdf/an200-4.pdf
« Last Edit: November 17, 2014, 11:34:42 am by nixxon »
 

Offline mweymarn

  • Contributor
  • Posts: 15
  • Country: de
  • Callsign: DL2VW
Re: HP 5316A Teardown and Frequency Standard
« Reply #14 on: November 16, 2014, 02:40:56 pm »
amc184 kindly sent me one of his PCBs. I used it to upgrade the timebase of my HP 5316B. This counter has a slightly different mechanical layout, nevertheless the PCB fits perfectly as can be seen in the photo.

Unified standoffs and spacers are hard to come by in Europe. Luckily standard M3 10mm standoffs fit perfectly. Although a little tight the M3 screws can be run through the 6-32 nuts which are riveted to the counter's main board.

The OCXO calibrates very well and is extremely stable. An amazing upgrade, especially compared to the standard timebase!
 
The following users thanked this post: Thilo78

Online Alex Eisenhut

  • Super Contributor
  • ***
  • Posts: 3550
  • Country: ca
  • Place text here.
Re: HP 5316A Teardown and Frequency Standard
« Reply #15 on: November 16, 2014, 05:36:24 pm »
Odd, mine looks like this

Hoarder of 8-bit Commodore relics and 1960s Tektronix 500-series stuff. Unconventional interior decorator.
 

Offline PA0PBZ

  • Super Contributor
  • ***
  • Posts: 5259
  • Country: nl
Re: HP 5316A Teardown and Frequency Standard
« Reply #16 on: November 16, 2014, 05:40:57 pm »
Odd, mine looks like this

Because it has the C input option.
Keyboard error: Press F1 to continue.
 

Online Alex Eisenhut

  • Super Contributor
  • ***
  • Posts: 3550
  • Country: ca
  • Place text here.
Re: HP 5316A Teardown and Frequency Standard
« Reply #17 on: November 16, 2014, 06:20:05 pm »
Odd, mine looks like this

Because it has the C input option.

Yeah but I mean graphically, even the HP logo isn't the same.
Hoarder of 8-bit Commodore relics and 1960s Tektronix 500-series stuff. Unconventional interior decorator.
 

Offline amc184Topic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: nz
Re: HP 5316A Teardown and Frequency Standard
« Reply #18 on: November 17, 2014, 05:22:00 am »
Yeah, yours is a little bit older, maybe 1983.  Mine is around 1985 vintage.

Thanks for the picture mweymarn, it looks really good, nice clean build.  I checked out the service manual as best I could when I designed the PCB to try to make sure it would fit both types of 5316, I'm glad it worked.
 

Offline nixxon

  • Frequent Contributor
  • **
  • Posts: 342
  • Country: no
Re: HP 5316A Teardown and Frequency Standard
« Reply #19 on: November 18, 2014, 01:17:37 pm »
It looks like the PCB will work.  The pinout for the earlier 5315 was:

1) NC
2) RF
3) GND
4) VCC (5V regulated)
5) NC
6) NC

So it looks like your later 5315 is designed to allow the use of either type of module.  I'll PM you about getting a PCB shipped to you.  I've also attached the schematic, design files and gerber.  Anyone should feel free to use them to make their own PCBs, or modify them to suit their own counter.  They're made with DipTrace, the type numbers in the bill of materials are element14 part numbers.

Thank you for attaching the schematics, layout files and gerber. They gave me a reason to download DipTrace, simply to be able to read all the files. It looks very impressive. Maybe, the layout files will make a good starting point to learn a little about DipTrace as well.

When I look at the schematics ("004 B1.pdf"), I notice that there are 2 capacitors (C101 and C102) in parallel between +10 VCC (pins 1+6) and GND (pin3). C101 is huge, 330 uF, compared to C102, 100 nF. They both add up to a capacitance of 330uF + 0.1uF = 330.1uF.

My noob guess is that C101 is there to reduce the ~600mVpp ripple of the VCC.

My noob question is: What is the purpose of C102?

(I notice a similar thing between the +5V output of U101 (LM1085, powering the OCXO) and GND as well. One huge, 330uF capacitor, C105, in parallel with multiple tiny capacitors, C104, C106 and C107. They all add up to 330.201uF. What is the purpose of adding 0.201uF to 330uF?)
 

Offline SharpEars

  • Supporter
  • ****
  • Posts: 634
  • Country: us
Re: HP 5316A Teardown and Frequency Standard
« Reply #20 on: November 18, 2014, 02:25:36 pm »
Perhaps a stupid question but:

If you already have a two channel oscilloscope with a built in frequency counter as one of its many stat/math functions, is there any point in getting a frequency counter that covers the same range as the scope?

In other words, why are all of you getting and/or in possession of these if modern scopes give you frequency counts along with waveforms?
 

Offline nfmax

  • Super Contributor
  • ***
  • Posts: 1624
  • Country: gb
Re: HP 5316A Teardown and Frequency Standard
« Reply #21 on: November 18, 2014, 02:51:02 pm »
Here's a few reasons!
  • My counter goes up to 3GHz, but my fastest scope is only 200MHz
  • The counter has interpolation for more digits/second resolution
  • With a universal counter, you can measure things like the number of pulses on an interrupt line that occurred while /RESET was low
 

Offline amc184Topic starter

  • Regular Contributor
  • *
  • Posts: 126
  • Country: nz
Re: HP 5316A Teardown and Frequency Standard
« Reply #22 on: November 18, 2014, 06:38:19 pm »
Nixxon; the small capacitors are for decoupling.

SharpEars; A few reasons.  I mainly leave this counter connected to the sync input of my signal generator (HP 3312A with no frequency display).  Sometimes I may be viewing something other than an amplifiers output, such as its supply rails, yet I still want to know the frequency of what I'm feeding it.  Also for accuracy.  I can't discipline my 'scope to a 10MHz reference.

Aside from those concerns, a counter is cheap.  I think the 5316A cost me about USD30, maybe another USD20 for the clock module.  You only need it to be useful now and again to justify that sort of cost.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: HP 5316A Teardown and Frequency Standard
« Reply #23 on: November 18, 2014, 07:00:21 pm »
Aside from those concerns, a counter is cheap.  I think the 5316A cost me about USD30, maybe another USD20 for the clock module.  You only need it to be useful now and again to justify that sort of cost.

The HP5316A is also a reciprocal counter so it returns a constant number of digits at a given gate time producing high resolution at low input frequencies.  On analog oscilloscopes which support gated measurements, reciprocal counters are common but it is not clear if that is the case with DSOs.

DSOs with long record lengths should be able to equal the performance of a good timer/counter without additional hardware support but I have yet to see this happen.  They all seem to be mediocre at best.
 

Offline nixxon

  • Frequent Contributor
  • **
  • Posts: 342
  • Country: no
Re: HP 5316A Teardown and Frequency Standard
« Reply #24 on: November 19, 2014, 12:44:59 pm »
Nixxon; the small capacitors are for decoupling.
(...)

Thank you for making that clear.

Is there any reason why the 5V OCXO can not be powered directly by the existing +5.00V supply (pin 4)?

To me it seems that the instruments's +5.00V supply (pin 4) is powered by the unregulated +7,5V that is also present on pins 1 and 6.

The Morion MV85 OCXO has a peak power consumption of <600 mA during warm-up.
« Last Edit: November 19, 2014, 02:26:36 pm by nixxon »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf