Author Topic: No luck on my first attempt at acquiring a cesium frequency standard  (Read 41681 times)

0 Members and 1 Guest are viewing this topic.

Offline hydrogen maser

  • Regular Contributor
  • *
  • Posts: 129
  • Country: us
So I bought a Datum 4040A Cesium Frequency Standard, just arrived today. I figured given the price (very cheap) that there was a 98% chance that it did not work. And yep, plug it into AC and it immediately starts making that typical failed switch mode power supply noise. I thought great! Hopefully all I have to do is fix the power supply! Getting a device with a failed power supply is often an easy and cheap fix. No such luck. One of the circuit boards has impact damage to it, it is physically broken. There is no damage to the case top so my guess is that what happened was it was opened up for troubleshooting and repair and some idiot dropped something heavy in it and broke that circuit board. Oh well, I knew it was a lottery shot that I would get something this cheap that was fixable, but boy is it disappointing to see something like that. I highly doubt I will be able to find a replacement circuit board.

Edit: SORRY! I just realized that this post is of no use to anyone and it is just me sharing my tale of woe, so does anyone know where I might get parts for a 4040A?
« Last Edit: March 09, 2016, 12:00:58 am by hydrogen maser »
 

Offline uncle_bob

  • Supporter
  • ****
  • Posts: 2441
  • Country: us
So I bought a Datum 4040A Cesium Frequency Standard, just arrived today. I figured given the price (very cheap) that there was a 98% chance that it did not work. And yep, plug it into AC and it immediately starts making that typical failed switch mode power supply noise. I thought great! Hopefully all I have to do is fix the power supply! Getting a device with a failed power supply is often an easy and cheap fix. No such luck. One of the circuit boards has impact damage to it, it is physically broken. There is no damage to the case top so my guess is that what happened was it was opened up for troubleshooting and repair and some idiot dropped something heavy in it and broke that circuit board. Oh well, I knew it was a lottery shot that I would get something this cheap that was fixable, but boy is it disappointing to see something like that. I highly doubt I will be able to find a replacement circuit board.

Edit: SORRY! I just realized that this post is of no use to anyone and it is just me sharing my tale of woe, so does anyone know where I might get parts for a 4040A?

Hi

I guess the first question would be "how much did you pay?"

The second question would be "can you return it?" (eBay is pretty good about this if it was not properly described).

The most likely case:

You have a unit made up out of the scraps from somebody else repairing a dozen units. Yours is physically complete, but unlikely to work without a lot of effort. Until you get *all* of the various issues sorted out, there is no easy way to check the beam current in the tube.  If the tube is shot, you have a very large paperweight. If the tube is running way low on current, you have a standard that may not be as good as some OCXO's out into the > 10K second region.

You might consider a cheap GPSDO as an option ....

Bob

« Last Edit: March 09, 2016, 12:38:19 am by uncle_bob »
 

Offline hydrogen maser

  • Regular Contributor
  • *
  • Posts: 129
  • Country: us
No, it is not returnable and not from ebay and I paid very little for it and I knew that it was a total lottery ticket. It does not look like it was pieced together from random parts. The lack of tool marks on all the RF connectors and screws (none missing) and other connectors and the consistent wiring ties and just the general look of a factory job leads me to believe that this is an original unit and not a Frankenstein job. Nothing about it looks like anyone has touched it since the day it was manufactured. The tube seal says 1999 so depending on when it went out of service, is the tube any good now? Who knows. But given the low cost to me I am willing to take a gamble on spending a little to try to fix it or at least have some fun with it.   
« Last Edit: March 09, 2016, 12:53:37 am by hydrogen maser »
 

Online Jay_Diddy_B

  • Super Contributor
  • ***
  • Posts: 2525
  • Country: ca
Hi,

May be you can post some pictures?

I have a 5061A that was once of the US Naval Observatory.

Regards,

Jay_Diddy_B

 

Offline uncle_bob

  • Supporter
  • ****
  • Posts: 2441
  • Country: us
No, it is not returnable and not from ebay and I paid very little for it and I knew that it was a total lottery ticket. It does not look like it was pieced together from random parts. The lack of tool marks on all the RF connectors and screws (none missing) and other connectors and the consistent wiring ties and just the general look of a factory job leads me to believe that this is an original unit and not a Frankenstein job. Nothing about it looks like anyone has touched it since the day it was manufactured. The tube seal says 1999 so depending on when it went out of service, is the tube any good now? Who knows. But given the low cost to me I am willing to take a gamble on spending a little to try to fix it or at least have some fun with it.

Hi

The Datum tubes seem to have about a 10 to 12 year life. Most of these go into service within 6 months of manufacture.

Bob
 

Offline wraper

  • Supporter
  • ****
  • Posts: 13174
  • Country: lv
How complicated is that PCB? Is it multilayer? As for me, I might just glue the pieces together and connect together the broken tracks. As why this damage could happen, IMO it was dropped or received mechanical damage during the shipping. If the board is heavy and flimsy enough, it can break just because of the acceleration shock.
 

Offline Macbeth

  • Super Contributor
  • ***
  • Posts: 2562
  • Country: gb
1990's PCB is probably double sided, maybe 4 layer at worst. It should be easy to repair with some solder wire bridges etc. if just physical impact damage on some corner of a board. I fixed many a VCR that the wife chucked at the hubby back in the early '90s - when she found him watching blue movies on VHS. The tape would be trapped in the machine of course and we had to return sans such filth. But with a nicely bodged PCB with wire links soldered to scraped off solder mask tracks.
 

Offline hydrogen maser

  • Regular Contributor
  • *
  • Posts: 129
  • Country: us
pics:





Impact damage:






 

Offline wraper

  • Supporter
  • ****
  • Posts: 13174
  • Country: lv
Where is the crack? I only see a conformal coating.
 
The following users thanked this post: Kean

Offline uncle_bob

  • Supporter
  • ****
  • Posts: 2441
  • Country: us
Hi

That's a pretty major crack in the board. It's also pretty un-likely that it's a two layer board. No traces on the top side for a board that heavily loaded .... mulit layer.

Best bet ... hope the cheap deal on this one comes around again.

Bob
 

Offline wraper

  • Supporter
  • ****
  • Posts: 13174
  • Country: lv
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #10 on: March 09, 2016, 01:45:05 am »
Hi

That's a pretty major crack in the board. It's also pretty un-likely that it's a two layer board. No traces on the top side for a board that heavily loaded .... mulit layer.

Best bet ... hope the cheap deal on this one comes around again.

Bob
Still don't see the crack, only the edge where the conformal coating ends.
 
The following users thanked this post: Kean

Offline hydrogen maser

  • Regular Contributor
  • *
  • Posts: 129
  • Country: us
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #11 on: March 09, 2016, 01:57:14 am »
Hi

That's a pretty major crack in the board. It's also pretty un-likely that it's a two layer board. No traces on the top side for a board that heavily loaded .... mulit layer.

Best bet ... hope the cheap deal on this one comes around again.

Bob

Yeah, I'm pretty much screwed on this one. Looks like it is a 4 layer board and there is no way I can fix it. Best bet it is look for another one for cheap and hope the parts I need from it match up to mine. Who knows, might even get a good tube from another broken one. Oh well, that is how it goes. Sometimes you win, sometimes you don't.
 

Offline uncle_bob

  • Supporter
  • ****
  • Posts: 2441
  • Country: us
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #12 on: March 09, 2016, 02:02:32 am »
Hi

That's a pretty major crack in the board. It's also pretty un-likely that it's a two layer board. No traces on the top side for a board that heavily loaded .... mulit layer.

Best bet ... hope the cheap deal on this one comes around again.

Bob

Yeah, I'm pretty much screwed on this one. Looks like it is a 4 layer board and there is no way I can fix it. Best bet it is look for another one for cheap and hope the parts I need from it match up to mine. Who knows, might even get a good tube from another broken one. Oh well, that is how it goes. Sometimes you win, sometimes you don't.

Hi

There still seems to be a bit of confusion between what is damage and what is conformal coating on the board.

Bob
 

Offline wraper

  • Supporter
  • ****
  • Posts: 13174
  • Country: lv
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #13 on: March 09, 2016, 02:09:19 am »
Call me blind but I cannot see the crack.
 

Offline hydrogen maser

  • Regular Contributor
  • *
  • Posts: 129
  • Country: us
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #14 on: March 09, 2016, 02:10:36 am »
Hope this helps point out why I am screwed on this one :) That is a crack in the board - trust me.

« Last Edit: March 09, 2016, 02:12:11 am by hydrogen maser »
 

Offline uncle_bob

  • Supporter
  • ****
  • Posts: 2441
  • Country: us
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #15 on: March 09, 2016, 02:13:59 am »
Hope this helps point out why I am screwed on this one :) That is a crack in the board - trust me.



Hi

Ok, that's how I was interpreting the pictures. Looking at the angle shot, it's very easy to decide that's conformal coat. Since you can see what it is, I'm sure it's apparent that the board is cracked.

Bob
 

Offline hydrogen maser

  • Regular Contributor
  • *
  • Posts: 129
  • Country: us
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #16 on: March 09, 2016, 02:16:39 am »
Sorry if the pic was not clear, my camera sucks. What I outlined in blue is not a shadow edge from a coating or anything else, it is a very nasty crack in the board.
 

Offline uncle_bob

  • Supporter
  • ****
  • Posts: 2441
  • Country: us
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #17 on: March 09, 2016, 02:22:07 am »
Sorry if the pic was not clear, my camera sucks. What I outlined in blue is not a shadow edge from a coating or anything else, it is a very nasty crack in the board.

Hi

That board has a *lot* of stuff on it.

Have you pulled the board yet? It's not clear if there is something under that board or not.

Bob
 

Offline hydrogen maser

  • Regular Contributor
  • *
  • Posts: 129
  • Country: us
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #18 on: March 09, 2016, 02:25:01 am »
Another angle:

 

Offline wraper

  • Supporter
  • ****
  • Posts: 13174
  • Country: lv
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #19 on: March 09, 2016, 02:27:22 am »
I don't get how the board can be cracked this way, and the most of it, how the crack can match the edge of the conformal coating by 100%. Not a single part is damaged too.
 

Offline wraper

  • Supporter
  • ****
  • Posts: 13174
  • Country: lv
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #20 on: March 09, 2016, 02:28:43 am »
Another angle:


No crack, only conformal coating. If you don't trust me, remove the board and check the other side for the crack being present.
 

Offline hydrogen maser

  • Regular Contributor
  • *
  • Posts: 129
  • Country: us
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #21 on: March 09, 2016, 02:38:09 am »
No crack, only conformal coating. If you don't trust me, remove the board and check the other side for the crack being present.

HA! You are correct! That is not a crack, looked like it to me from many angles until I got the microscope out to look at it. My eyesight is not the same as it was 30 years ago. Thank you! Well, this changes things quite a bit. Back to the PS repair! Always glad to be proven wrong! :) 
 

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 10668
  • Country: gb
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #22 on: March 09, 2016, 03:04:06 am »
Hmmmmm, nasty crack but eminently repairable in my experience.

You would need to desolder the components that bridge the crack. Then complete the crack 'circle' with a knife and remove the 'island' complete.

Using a microscope, inspect the edges of the island looking for the tracks. Sometimes the edge needs to be cleaned up with a light sanding. A Dremel multitool may also be used to step down through the layers at the edge of the island. Once all PCB tracks have been identified, the dremel tool is used to reveal them enough for soldering Kynar jumper wires. The same is done to the main PCB so that the Kynar jumpers can straddle the gap between it and the island. It is important to slightly raise the island above the main PCB, like a daughter board to enable the Kynar wires to be soldered around the edges. A support structure will need to be constructed but that is purely a mechanical challenge. Once all wire links are in place, the island may be lowered to within a couple of mm of the main PCB where is needs to be fixed in place using epoxy or some such mechanical fixing method. The jumper wires just sit underneath. The components that stradle the crack will need to be adjusted or modified to fit the slightly elevated island. Wire wrap sockets with their cong legs may be used for chips. Components with longer leads may be required or SMT replaced with through-hole technology.

An alternative approach once the various PCB tracks have been identified is to prepare the island and main PCB edges and attempt to link the tracks using very short lengths of thin tinned copper wire. This can be challenging depending upon the layout of the multi layer tracks.

If the multi layer tracks are too close together at the edge of the PCB, you can use continuity checks to locate their connection points on components. The jumper wire may then be connected to the component leads instead of to the PCB tracks. In some cases this is a far better approach to the problem. You end up with somewhat of a spiders web of Kynar wire, but it works.

This sort of repair is the sort of thing I used to do for the fun of the challenge...... People said the PCB's were beyond repair...... I proved that to not be the case. Some equipment I repaired had been deliberately damaged wiih a hammer blow through the middle of the main PCB. Unrepairable ? No, provided no unique or custom chips had been destroyed. If you really want to wreck kit, you smash up the silicon :)

I might add that this sort of repair is reserved for non mission critical applications such as data recovery or, as in your case, a desire to bring a unit back from the dead.

Another hint, I used to buy brass tube, rod  and hollow box section from the modelling shop. I used it to create temporary or permanent scaffold for the island PCB The brass solders easily and is pretty rigid. I used the brass to reinforce the epoxy joining of the PCBs by so,dering to the ground plane copper. I never had a repair fail and the PCB strength was maintained.

One for a rainy day maybe ?

Fraser
Cogito, ergo sum
 

Offline Fraser

  • Super Contributor
  • ***
  • Posts: 10668
  • Country: gb
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #23 on: March 09, 2016, 03:15:33 am »
Having just looked up the Burr Brown PCM54 chip, I see that it is just a Digital to Analogue converter, and the MC6821 is the very common PIA that many MC6800 series processors used.
The data sheets for the two chips will give you some idea of the connectivity and may help identify what is likely connected to other parts of the PCB, such as the address and data lines. They may be jumper end to other chips that  need the same, or nearby buffer chips that serve them, such as the common 74LS245.

Fraser
Cogito, ergo sum
 

Offline Vgkid

  • Super Contributor
  • ***
  • Posts: 2619
  • Country: us
Re: No luck on my first attempt at acquiring a cesium frequency standard
« Reply #24 on: March 09, 2016, 03:23:00 am »
Best of luck, what ocxo does it have inside of it(should be decent).
If you own any North Hills Electronics gear, message me. L&N Fan
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf