HP 5385A

[metrologist]

Are these counters good? I wonder why I find so few posts on here about them. Is there anything in particular to know about them? (I'm really trying to stay away from TheSteve's sale thread...)

I made a little low offer on one (no options) from ebay and to my surprise they accepted it! 

[Muxr]

They are great. This one is mine after I fixed it:



Some of them have an option with an internal battery which I imagine would be really handy for HAM operators on the go. LCD based display means you don't have to worry about screen burn in so you can leave them on all the time. Say as a companion to an analog scope via ext_trigger.

[metrologist]

Your unit looks very clean. Mine arrived and is just a little scuffed up, but good condition. I've read that the LCD's can bleed, but our's are still good.

I know the difference between accuracy and precision, which is more like stability. The manual states an accuracy specification of ~+/-7.72mHz, if I am interpreting it correctly.
(using 10MHz and 1E-7/mo aging rate over 10s gate)

I let it warm up for a few hours and measuring my OCXO I am seeing the readings drifting around within a 12mHz window. I'm not sure if that is expected, but thought I would see that last digit toggle between just two values.

It might be my setup contributing as it's just a loose oscillator connected to my linear PS with a 2k trimpot on the tuning pin.

I haven't found any interesting features like my HP5314A.

(except I just saw the DC input jack. This is good because I imagine one could easily use it with an Arduino to log measurements out in the field) 

[metrologist]

Maybe I'm not interpreting the accy spec correctly. I rethink the calculation and get something much less.

I can say that my 5314A with the TCXO option seems more stable, to the resolution it can measure. The 5385A started moving around a lot more so I was not sure if my OCXO was drifting or the counter.

Then I pulled my second oven and synched the two OCXOs using my scope to beat them together, and there was still a couple mHz back and forth over the day.

[metrologist]

It's drifting more now. I see almost a full hurt over a couple days and a hundred or so milihurt short term. 

This I cannot tolerate, I will have to procure an OCXO. It looks like there are some additional changes to the counter board, not just a plug in swap, and I have not found the actual HP oscillator anyway. Or I wonder if it could be just a noisy tune circuit?

Dave bought a Chinese reference for his higher model. Anyone know where he got it or something reasonable?

I don't think I can really complain to the seller. They said it came in a lot, but they had it connected and mentioned it looked like a recent cal because it was very close to the input frequency. This is not a high volume seller.

[metrologist]

I just ordered a CTI OCXO 5V from ebay. I do not see a tuning pin marked on it though. It has to be tunable, right?

Also, in the HP5385A schematic, I do not see the tuning circuit anyway. S2 just goes to the 5V power switch. Are these tuned with the supply voltage? 

[TorbenRune]

I have a similar experience with a HP 5385A with option 004 (oven heated oscillator). After testing I could see (by comparing it to another OCXO in my HP8656B) that it was drifting with temperature, indicating that the heater element is probably defect. Touching the OCXO while powered on, confirms this – it has the same temperature as the surroundings.

This week I received a GPS disciplined 10 MHz frequency reference (Symmetricon TrueTime 600-00 XL-AK). Using that to trigger my scope, I clearly see the drift of the 5385A, while the 8656B is rock solid.

My plan is to use the Symmetricon as a common freq. ref. for all instruments, but that limit my usage of the 5385A to my bench, so I am considering changing the OCXO anyway.

metrologist, did your repair go as planned, and did you find a OCXO that actually fit into the 5385A? I assume, that the OCXO must be tunable, to be sure to get the high precision.

[metrologist]

I have not installed the oscillator shown, but that one will fit. I'll need to build it onto a daughter board with the tuning circuit, just have not had time. It was easier to use an external reference.

[AdamHi]

metrologist - from what i could glean, the ebay OCXO you have is a square wave output, is that correct, or am i misreading that? Does the HP counter do well with either?? The output from my 5385a internal TCXO is a sine wave (when sampled from the BNC connector on the rear).

I'm new at modifying HP equipment - any traps I should look out for if i'm going to try to install an OCXO internally?

--adam

[Miti]

I know this is more than half a year old but if anybody is interested, I still have 9 boards left.

https://www.eevblog.com/forum/testgear/hp5385a-ocxo-timebase/msg1802765/#msg1802765

[metrologist]

Sorry I missed your post, Adam. Still using an external reference, which is sine wave. I wouldn't think either is a problem.

Thanks for the lead, Tiki. The OCXO's for this board are available on ebay for $6USD from queen*s_land, who I bought mine from originally. In my photo, the tuning pin is upper left and I believe the upper right is the oven ready pin. Tiki, where did you attach the ref signal? I don't recall what the innards of the counter looked like, so is your board a drop-in replacement?

I 3D printed a dual wall plastic cover for this osc, which has 1/8" sealed air cavities around the sides. I need to make one bigger.

[Miti]

Sorry I missed your post, Adam. Still using an external reference, which is sine wave. I wouldn't think either is a problem.

Thanks for the lead, Tiki. The OCXO's for this board are available on ebay for $6USD from queen*s_land, who I bought mine from originally. In my photo, the tuning pin is upper left and I believe the upper right is the oven ready pin. Tiki, where did you attach the ref signal? I don't recall what the innards of the counter looked like, so is your board a drop-in replacement?

I 3D printed a dual wall plastic cover for this osc, which has 1/8" sealed air cavities around the sides. I need to make one bigger.

If Tiki = Miti, I replied in the other thread. 

[metrologist]

LoLs, there isn't even a member named Tiki on this board. Sorry about that. Will go read your post.

[Housedad]

I know this is more than half a year old but if anybody is interested, I still have 9 boards left.

https://www.eevblog.com/forum/testgear/hp5385a-ocxo-timebase/msg1802765/#msg1802765

How do I purchase one from you?

[Miti]

I know this is more than half a year old but if anybody is interested, I still have 9 boards left.

https://www.eevblog.com/forum/testgear/hp5385a-ocxo-timebase/msg1802765/#msg1802765

How do I purchase one from you?

Here's the schematic.
Capacitors are ceramic X7R 0603 except 10uF that are 1206 and L1 is a 0805 ferrite bead.
MAX6126AASA has various output voltages. Choose the one that fits your need.
R1, R2, R3 must be calculated for your OCXO. There's a bridge under R1 that you can cut if needed.
Some Digikey part numbers:
FERRITE BEAD 835 OHM AXIAL 1LN     240-2491-ND
FERRITE BEAD 220 OHM 0805 1LN      445-172905-1-ND
CAP CER 10UF 10V X7R 1206              445-1602-1-ND
TRIMMER 100 OHM 0.5W PC PIN TOP   3296W-101LF-ND

L2, C9, C10 was intended to be a low pass filter but first of all I missed a resistor at the output of the OCXO and second, it is not needed, just install a 10 Ohm resistor or even a jumper.

[Housedad]

Miti, do you know where I can download a datasheet on the CTI OCXO you used?   I can't seem to even find the maker CTI, much less the datasheet.

[Miti]

Miti, do you know where I can download a datasheet on the CTI OCXO you used?   I can't seem to even find the maker CTI, much less the datasheet.

This is a seller that points to a datasheet:
https://www.ebay.com/itm/1-USED-CETC-CTI-OSC5A2B02-10MHz-5V-26-26-12-Square-Wave-OCXO-Crystal-Oscillator/332389156868?hash=item4d63f00804:g:5VIAAOSwruVZxm6k:rk:2:pf:0

And it points here:

http://www.cti.ac.cn/products/system/professional/crystal/2013/0725/289.html

[Housedad]

Thanks again!!! 

[DC1MC]

Got my board form Miti and in the same day the OCXOs from China,  , the board looks nice and ALL the OCXO are outputting a nice square wave of 10.0000 MHz  , the scratches are exagerated by the lightning and are superficial, the seller is to be recommended, she even put a small gift in the box.
Now while I'll get the rest of the BOM, I have to (finally) start either my Rubidium or the GPSDO boards to select one of the OCXO.
BTW, the board looks very nice and professionally done, it arrived from CA to DE in reasonable short time as a postcard letter  , very nicely packed  , so congrats to Miti and I'll let you know what was the performance of the best OCXO and how my board looks assembles.
Also, soon to have three 10MHz OCXO available for DE/EU members  .

 Cheers,
 DC1MC

EDIT: The story will continue here: https://www.eevblog.com/forum/testgear/hp5385a-ocxo-timebase/
 

[Housedad]

finally ordering the small parts after my OCXO came in.  What is the difference between the two part numbers on digikey?

They show this:

MAX6126AASA21+-ND

and

MAX6126AASA21+

What is the difference that the -ND makes? 

[Kean]

What is the difference that the -ND makes?

Don't know if it is true, but I was told it stands for "NO DISCOUNT" - just a Digikey part numbering thing.   

DK often use interesting variations on the manufacturers part numbers, probably for internal reasons, or just being lazy... although it seems like extra work!

[Miti]

Please see this for DK part #

https://forum.digikey.com/t/what-do-the-digi-key-suffixes-tr-ct-dkr-and-nd-mean/30

[Housedad]

Thanks, MIti.  I was definitely confuzelled on that one.