No luck on my first attempt at acquiring a cesium frequency standard

[edpalmer42]

I have his big brother, the 4065A.  The main board looks identical to yours.  It also has the internal cesium unit model #5045A including the strange conformal coating over part of the board.  You can see pictures of mine here:

http://s701.photobucket.com/user/edpalmer42/media/Datum%20FTS-4065A/SAM_0322_zps0r3j6ssc.jpg.html?sort=4&o=0

But before you get too involved, I have two things to tell you about.

First, a cesium standard has a limited lifetime.  Whenever it's on, it's using up its supply of cesium.  Once that's gone, the cesium tube is a paperweight and the entire unit is a boat anchor.  So, if you've got the equipment, you should test the tube.  I've attached the test results from my tube.

Second, on the main board, look at U51.  It's in a PLCC44 package.  It's an STEL-1173 NCO (Numerically Controlled Oscillator).  This chip almost always fails with age.  Since it's a obsolete, specialized chip with no second source, it's hard to find.  I got mine a couple of years ago from here:

http://www.aliexpress.com/item/Free-shipping-10PCS-STEL-1173-CM-PLCC/2023682701.html

I chopped out the old chip and soldered in a socket so I can replace it easily if necessary.

Actually, the chip failure could be a good thing.  Hopefully, the chip failed before the tube ran out of cesium.  That's what happened with mine.  There's no way to tell how much cesium is left, but mine is working okay for now.

So, if the power supply repairs look reasonable, go for it.  It'll make the other tests a lot easier.  If not, maybe you could remove the failed power supply and substitute external supplies as necessary.

Let me know if (when!) you have more questions.

Ed

[hydrogen maser]

Just so everyone is clear on who is the idiot in this situation, that would be me. I apologize for not recognizing my failing eyesight. I often rely on cameras and lenses to help me see things as I am far from being young anymore. The "crack" in the board is no such thing. There is a conformal coating on the board that in some small defense of my stupidity I would note that it is slightly odd and nothing like anything I have seen before, just a single small spot on a large board. edpalmer42, thank you for your information and advice. 

[hydrogen maser]

Looks like I am at a dead end on this one for now. The main switch mode power supply appears to be working ok, it looks like something is wrong with the high voltage power supply in the 5045A. The unit won't go into the initialization routine when it is brought out of storage mode. At 2KV DC, I don't have the equipment or courage to troubleshoot it. Not worth messing around with that kind of voltage. There is also evidence that someone else has been in there because the RF connector for the 10mhz oscillator was disconnected and one of the mounting brackets for the tube assembly was crooked.

[Fraser]

You are a very honest person for admitting your error in misidentifying a crack. Respect to you sir.

Regarding eyes.... I am 48 and must use magnifying devices when working on electronics. It has taken me quite some time to come to terms with my eyes ageing. They used to be 20:20 with excellent near vision. Now I find I need to hold items at arms length, or wear my glasses.

Fraser

[uncle_bob]

Looks like I am at a dead end on this one for now. The main switch mode power supply appears to be working ok, it looks like something is wrong with the high voltage power supply in the 5045A. The unit won't go into the initialization routine when it is brought out of storage mode. At 2KV DC, I don't have the equipment or courage to troubleshoot it. Not worth messing around with that kind of voltage. There is also evidence that someone else has been in there because the RF connector for the 10mhz oscillator was disconnected and one of the mounting brackets for the tube assembly was crooked.

Hi

... so back to the "maybe a kit of parts" theory.

Bob

[edpalmer42]

Do you have the manual and have you hooked a computer to the RS-232 port?  Do you have the old DOS monitor program?  What responses are you getting?

In case you know this, I'm playing to the audience!  When you power up a cesium standard that's been sitting around for a long time, it won't start right away.  A Cesium tube requires a really hard vacuum so the first thing it does is pump down the tube to an acceptable value using an ion pump.  This could take a day or more - even a week or more, depending on the state of the tube.  The ion pump runs in storage mode, but you have to turn the unit on to query it.  There's also a monitor point on my ion pump power supply that gives 1mv=1ua.  You could measure this in storage mode.

Ed

[edpalmer42]

Here's some info to get you started.  The manual is for the 4040B.  There might be a few changes, but the majority will be good.

The monitor program runs under DOS.  It might run in a VM.  It's kind of a dopey program so you might want to just hook a computer to the RS-232 port and use the info in the manual to type the command strings through a terminal program.  I've done both.  Your call.

To monitor the ion pump, look at the two high voltage supplies.  Remember that you can do this in either storage or operate mode.  One supply has three low-voltage terminals on the side.  Power, ground, and monitor.  Measure the voltage between ground and monitor.  You should see a negative voltage of a few hundred millivolts or less.  What do you see?

Oops!  Turns out you can't attach a file greater than 1 Meg.  I've uploaded the manual to Mediafire.  Help yourself.

http://www.mediafire.com/download/6icdc7v35c68t16/4040B_mnl.pdf

Ed

[hydrogen maser]

Thank you edpalmer42! That helps a lot. This is a very strange device with some odd operational behavior. Now that I have a manual it makes more sense. Of course in my box of old serial port cables and gender changers I have everything except the the parts I need so I ordered the right cable, should be here early next week. From reading the manual it still looks like something is wrong because it should have gone into initialization mode regardless of how long it had been unplugged, but it does not. Of course, I have been wrong about everything about this device from day one so who knows what is going on with it. Once I have the right cable I will be able to see what (if anything) comes out of the serial port and that will tell me something. The disconnected oscillator cable was odd, but it might not have been human hands that did that. It has no tool marks on it and it fits loosely (unlike the nice tight fit of the other RF connectors), is on the bottom of the module, and has other RF cables that push against it. Might just be one of those things. I don't know. I thought it would be fun to play with this thing even if it did not work and I could at least take the 10mhz oscillator out of it for something else, but it has been more of a challenge than I thought it would be. It is very odd in a lot of ways, but in for a penny, in for a pound.

[guido]

Having just looked up the Burr Brown PCM54 chip, I see that it is just a Digital to Analogue converter, ....

Fraser

Yes a 16 bit stereo DAC for audio purposes. But not 'just' a DAC as this one has parallel inputs (same for the 55, anything else is serial). Hard to get if you need one and i did, as it is used in a calibration jig for an Iwatsu scope.

[Vgkid]

I figured if the tube was shot, it should contain a pretty decent 10Mhz ocxo.

[hydrogen maser]

Vgkid, yeah, obviously it has a 10mhz ovenized oscillator, don't know who made it as it is marked with a Datum part number and label. I assume it must be pretty decent tough.

[edpalmer42]

I figured if the tube was shot, it should contain a pretty decent 10Mhz ocxo.

The OCXO appears to be either a clone or an OEM version of the HP 10811.  Some other Cesium standards from FTS/Datum included a much fancier OCXO like the Datum 1000B in the FTS-4060.  Would love to find one of those.  Of course, since I'm a cheap S.O.B., that's not going to happen. 

Ed

[edpalmer42]

Thank you edpalmer42! That helps a lot. This is a very strange device with some odd operational behavior. Now that I have a manual it makes more sense. Of course in my box of old serial port cables and gender changers I have everything except the the parts I need so I ordered the right cable, should be here early next week. From reading the manual it still looks like something is wrong because it should have gone into initialization mode regardless of how long it had been unplugged, but it does not. Of course, I have been wrong about everything about this device from day one so who knows what is going on with it. Once I have the right cable I will be able to see what (if anything) comes out of the serial port and that will tell me something. The disconnected oscillator cable was odd, but it might not have been human hands that did that. It has no tool marks on it and it fits loosely (unlike the nice tight fit of the other RF connectors), is on the bottom of the module, and has other RF cables that push against it. Might just be one of those things.

Which, what, who?!!  Disconnected cables?!  You didn't say anything about that.  Loose fit?  Shouldn't be!  It's a standard SMA that should be tight enough that you need a small wrench to undo it.  Anything else that you haven't mentioned?

You haven't given us a picture of the bottom of the unit that has all the RF and Cesium parts.  Maybe you should do that.

Have you had time to make the measurement on the ion pump monitor?

Ed

[hydrogen maser]

Which, what, who?!!  Disconnected cables?!  You didn't say anything about that.  Loose fit?  Shouldn't be!  It's a standard SMA that should be tight enough that you need a small wrench to undo it.  Anything else that you haven't mentioned?

You haven't given us a picture of the bottom of the unit that has all the RF and Cesium parts.  Maybe you should do that.

Have you had time to make the measurement on the ion pump monitor?

Ed

Actually, I did mention that one of the RF connectors was disconnected, it was P51 in the image below (not screw on - a snap on-press fit connector): 



Full view of the bottom of the 5045A unit (mine does not look like the older ones I have seen on the web):

[uncle_bob]

Thank you edpalmer42! That helps a lot. This is a very strange device with some odd operational behavior. Now that I have a manual it makes more sense. Of course in my box of old serial port cables and gender changers I have everything except the the parts I need so I ordered the right cable, should be here early next week. From reading the manual it still looks like something is wrong because it should have gone into initialization mode regardless of how long it had been unplugged, but it does not. Of course, I have been wrong about everything about this device from day one so who knows what is going on with it. Once I have the right cable I will be able to see what (if anything) comes out of the serial port and that will tell me something. The disconnected oscillator cable was odd, but it might not have been human hands that did that. It has no tool marks on it and it fits loosely (unlike the nice tight fit of the other RF connectors), is on the bottom of the module, and has other RF cables that push against it. Might just be one of those things.

Which, what, who?!!  Disconnected cables?!  You didn't say anything about that.  Loose fit?  Shouldn't be!  It's a standard SMA that should be tight enough that you need a small wrench to undo it.  Anything else that you haven't mentioned?

You haven't given us a picture of the bottom of the unit that has all the RF and Cesium parts.  Maybe you should do that.

Have you had time to make the measurement on the ion pump monitor?

Ed

Hi

..... kit of parts .....Somebody tossed a bunch of "almost right" stuff together. You get a clone 10811 with a SMB connector and an SMA cable hanging loose.

Yes pictures / part labels / mounting locations would help sort things out a bit. Without that, the best guess is that this is a Frankenstein.

Bob

[hydrogen maser]

Not seeing what looks non-factory or wrong about it. Nothing looks like it was pieced together. Please point out exactly what looks wrong about it so I can understand your point because I don't get it.

[uncle_bob]

Not seeing what looks non-factory or wrong about it. Nothing looks like it was pieced together. Please point out exactly what looks wrong about it so I can understand your point because I don't get it.

Hi

Well, the pictures posted as my post was going up. The forum is a bit "interesting" in that respect. I just rejected this post as being "to soon". Since none of us know where the gizmo came from or anything like that, the only thing we can do is make random guesses.

Bob

[hydrogen maser]

I have been wrong about almost everything about this device but the one thing that I am most sure about is that there is no evidence that I have seen that this is anything but an original factory unit. I have been wrong about lots of stuff, this is a very strange beast, so take that for what it is worth (if anything), I am far from perfect. I have seen zero evidence anywhere inside this device that anyone has touched it since it was manufactured. My guess is that this particular era and revision of this very obscure device has just not ever been documented on the web yet. There is a first time for everything.

[edpalmer42]

Thanks for the picture.  Everything looks good.  I was worried that there might have been a missing module or something.  Yours has a different model of Cesium tube, but everything else looks pretty much identical to mine.

I had to go look at mine, but you are correct, it's an push-on SMB connector rather than a screw-on SMA.  Just goes to show the danger of relying on memory!  My apologies.  But you said that your connector is loose.  Mine has a nice 'snap' going on and requires a bit of effort to pull off.  Maybe do a close visual inspection on the inside of both ends.  There is another SMB connector nearby.  It's over toward the high voltage supplies.  Use that for comparison.

Bob, I don't see anything to suggest that this is a 'Frankenstein'.  If you compare HM's pictures to mine, the units look like twins as far as the Cesium/Main Board unit is concerned.  Makes sense since they're the same model number.  The basic module is 5045A.  Wrap different power supplies and user interfaces around it and you've got either HM's 4040A or my 4065A.  Yes, I know that Frankensteins happen, but only when you've got a bunch of identical units.  Who has a bunch of Cesium standards?? 

Ed

[uncle_bob]

Thanks for the picture.  Everything looks good.  I was worried that there might have been a missing module or something.  Yours has a different model of Cesium tube, but everything else looks pretty much identical to mine.

I had to go look at mine, but you are correct, it's an push-on SMB connector rather than a screw-on SMA.  Just goes to show the danger of relying on memory!  My apologies.  But you said that your connector is loose.  Mine has a nice 'snap' going on and requires a bit of effort to pull off.  Maybe do a close visual inspection on the inside of both ends.  There is another SMB connector nearby.  It's over toward the high voltage supplies.  Use that for comparison.

Bob, I don't see anything to suggest that this is a 'Frankenstein'.  If you compare HM's pictures to mine, the units look like twins as far as the Cesium/Main Board unit is concerned.  Makes sense since they're the same model number.  The basic module is 5045A.  Wrap different power supplies and user interfaces around it and you've got either HM's 4040A or my 4065A.  Yes, I know that Frankensteins happen, but only when you've got a bunch of identical units.  Who has a bunch of Cesium standards?? 

Ed

Hi

Well I can answer the last part of that:

These Cesium standards were sold in bulk to the telecom guys. They went into central office locations rather than into labs. Those guys never bought "just one". (No, that's not a guess, it's first hand knowledge). The fleet was deployed out around the country and the units went back to a depot to be maintained. The depot did the normal modern approach to fixing anything. They swapped out parts to get it working. As parts got scarce, they swapped parts between units to get ones working. The last unit on the shelf still had a full set of parts in it. Since these are high dollar items, you don't take them apart with a crowbar. Somebody tries really hard even on that last unit to get it running before giving up on it.

It doesn't have to look like a Frankenstein to be one. All it has to be is the parts donor that fixed the other dozen they had to repair.

Bob

[hydrogen maser]

I looked at the connector and cable end for P51 and they do not appear to be damaged. They look ok. Perhaps "loose" was the wrong word to describe it. How about "not super tight like all the other ones"? I could see given its position on the bottom of the unit (upside down) and with other cables that press on it how it could pop off in shipping. Not going to have a lot of time right now to play with this unit because I have a big work project going on this week. I am going to wait for the serial cable to arrive and then see what it shows (if anything). 

[edpalmer42]

Hi

Well I can answer the last part of that:

These Cesium standards were sold in bulk to the telecom guys. They went into central office locations rather than into labs. Those guys never bought "just one". (No, that's not a guess, it's first hand knowledge). The fleet was deployed out around the country and the units went back to a depot to be maintained. The depot did the normal modern approach to fixing anything. They swapped out parts to get it working. As parts got scarce, they swapped parts between units to get ones working. The last unit on the shelf still had a full set of parts in it. Since these are high dollar items, you don't take them apart with a crowbar. Somebody tries really hard even on that last unit to get it running before giving up on it.

It doesn't have to look like a Frankenstein to be one. All it has to be is the parts donor that fixed the other dozen they had to repair.

Bob

Which telcos did that?  You said 'deployed out around the country'.  Was that AT&T back in the day?  The reason I ask is that I worked in the Network Standards department of a Canadian Telco.  We had one of the twelve Class-1 Regional Switches in North America.  I don't know if that still means anything.  We certainly had Cesium Standards for synchronizing the digital network and operating in plesiochronous mode (geez I struggled with that word - now I can't forget it ) with other networks, but they were 48V powered, Central Office equipment.  I was very familiar with our labs and they had no lab-style Cesium Standards.  I think that was when I was first bit by the Time-Nuts bug although the term didn't exist back then.

Ed

[uncle_bob]

Hi

Well I can answer the last part of that:

These Cesium standards were sold in bulk to the telecom guys. They went into central office locations rather than into labs. Those guys never bought "just one". (No, that's not a guess, it's first hand knowledge). The fleet was deployed out around the country and the units went back to a depot to be maintained. The depot did the normal modern approach to fixing anything. They swapped out parts to get it working. As parts got scarce, they swapped parts between units to get ones working. The last unit on the shelf still had a full set of parts in it. Since these are high dollar items, you don't take them apart with a crowbar. Somebody tries really hard even on that last unit to get it running before giving up on it.

It doesn't have to look like a Frankenstein to be one. All it has to be is the parts donor that fixed the other dozen they had to repair.

Bob

Which telcos did that? 
Ed

Hi

My view of this is from the "supplier" end rather than the "user" end of things. At one point my boss was the head of FTS. Beyond that, I'd really rather not get into.

Pretty much all of them. You see Datum cesium's with KS numbers on them, you see them with Sprint tags, you see them with Verizon tags. That's just the ones I recognize. The whole "islands of sync" thing based on an autonomous local standard and interpolation at a large number of network edges is what seems to have driven it. HP went after the lab and military applications, but pretty much ignored the telecom's. Dataum / FTS saw that there was a very real market for network sync. That's what made them an attractive acquisition target for Symmetricom. 

Back in the day, AT&T was able to get away with a much more "top down" sort of approach. They were the only guys in town and they could propagate sync from a fairly narrow base of master standards. Once it all broke up, you had a bunch of "little Bell's" all that needed their own sync setup. Next along came the cellular guys and their needs for sync. Pile the explosion in network based telephony (and a bunch of people nobody has heard of hauling it) you have another empire of sync. Net result ... you need more master clocks.

Bob

[wkb]

Do the Rb standards used by telcos/cell operators fit in this same story somehow?  Cs being of a higher standard than Rb references obviously.

[edpalmer42]

Do the Rb standards used by telcos/cell operators fit in this same story somehow?  Cs being of a higher standard than Rb references obviously.

Yes and no.  Rb standards don't have an inherent end-of-life mechanism like Cs standards do.  An Rb standard could run for 20 years continuously.  But Rb standards do age while Cs standards don't.  So depending on your application, you might have to occasionally nudge a Rb standard to bring it back on frequency.  Typical aging rates for Rb are in the range of 5e-11 per month.  That's what they put in the spec sheets - YMMV.  It's not unusual to find a Rb standard that does much better than spec.  Cs standards are specified as having a maximum error as low as 1e-12 for life.

Ed