Author Topic: Datron 1062 (Read 33635 times)

gilbenl · « **on:** January 28, 2015, 10:47:57 pm »

Picked up this Datron 1062 off ebay to play with. The display looks not long for this world, but the price was cheap enough that it didn't matter. From what I can tell, the difference bw the 1061A and the 1062 is that the latter comes with ohms and AC RMS standard, but no current opt. Its really nice to work with even with an aging display. Unfortunately, a replacement is unobtainable AFAIK. I would be elated to hear it if someone suspects there may be a fix that doesn't require a new display. Pretty nifty inside so I figured I would share some pics. If anyone wants to see more, let me know.

Album: http://s76.photobucket.com/user/nick_gilbert2/slideshow/Datron%201062

lowimpedance · « **Reply #1 on:** January 28, 2015, 11:06:30 pm »

Nice clean unit, lots of nice old IC sockets to give trouble !, but also easy to lift the IC and re seat too.
The display's all seem to be affected in a similar way, perhaps it would be useful to at least check for ripple on the HT anode supply (ie dried out filter cap). Or just replace anyway.

Thanks for the pics.
If you do manage to fix post your solution here for others having similar problems.

Vgkid · « **Reply #2 on:** January 28, 2015, 11:09:17 pm »

The only source for those panaflex displays was ebay.

gilbenl · « **Reply #3 on:** January 28, 2015, 11:38:04 pm »

It could be a cap on the driver board, but I'm never that lucky. I know nothing about gas discharge displays, however, the diminished performance appears to be very uniform. It seems that the rate of degradation would vary if it were a matter of wear, or is that ignorance speaking? I've seen a couple "fail" units with good displays on ebay that don't seem to be moving. If I can get one for a good price, I may just cannibalize out the entire front panel and keep the guts for spare.

The unit is impeccable inside-not one speck of dust. And for the record, EVERY IC is socketed. I tried in vain to find one soldered directly to the board. That and the enclosed H configuration of the boards struck me as rather unique. Its in dire need of calibration so I'll definitely post some pics or video on the process. See how easy this "autocal" business is.

Mr.B · « **Reply #4 on:** January 29, 2015, 12:27:04 am »

I haven't seen one of those in years.
I used to laugh at the way the serial number was repeated all over the inside of the unit.
I wondered if they had allocated the serial numbers before testing the individual assemblies. Perhaps someone can shed some light on that theory.
In my "Push Buttons Box" at home I still have half a dozen of those rectangular push buttons with the red led - brand new, purchased as spares just in case.

EDIT: After a second, closer look, I see the serial numbers are all different. My bad memory - It was more than 25 years ago...

Mr.B · « **Reply #5 on:** February 24, 2015, 11:59:43 pm »

@gilbenl
Photo of buttons attached as per my PM.

dom0 · « **Reply #6 on:** February 25, 2015, 12:11:44 am »

I think I know these buttons. They are still available new... made by ITT cannon, I think. http://www.digikey.de/product-detail/de/SERL%20BK%20RD%207.62%20AU%20OA/401-1113-ND/417769

There are some hints (in German) about the 1061 here: http://www.amplifier.cd/Test_Equipment/other/Datron1061/Datron_1061.html

Regarding that aged VFD: There are some rejuvenation techniques around, see https://www.eevblog.com/forum/repair/vintageclassic-renovation-techniques/msg615992/#msg615992

HiTech · « **Reply #7 on:** February 25, 2015, 04:26:50 pm »

I had 2 Datron 1065A units and are very similar to yours. If yours has the internal battery for calibration memory, do not replace it unless you have external DC power applied across the terminals. So, keep the units plugged into live AC outlet otherwise it uses the battery for memory. Those are very good, near metrology-grade units. They can be calibrated but it's not a simple 1-2-3 process.

gilbenl · « **Reply #8 on:** February 25, 2015, 06:03:40 pm »

Quote from: HiTech on February 25, 2015, 04:26:50 pm

I had 2 Datron 1065A units and are very similar to yours. If yours has the internal battery for calibration memory, do not replace it unless you have external DC power applied across the terminals. So, keep the units plugged into live AC outlet otherwise it uses the battery for memory. Those are very good, near metrology-grade units. They can be calibrated but it's not a simple 1-2-3 process.

I appreciate the links, particularly to the german site. Shortly after starting this thread I purchased a second "broken" 1062 with the intent to harvest the display and keep in handy for spare parts. Seller said it showed "Error 6" on self test, which is a failure of the resistance circuit. Turns out the test was run in 4-wire mode. Turned that off and it passed. Bonus is that is has external input (ratio) and output, which will prove very handy to set up as a differential meter with a known standard. See picture below ("broken" on top of original"

So, now I have 2 functional 1062s (darn). The original still has a wonky display and I'm willing to be a bit more bold with it. I'm still hesitant to crank up the current on the display without having a definite "this works" protocol, namely because I don't know enough to not screw up. However, another resource suggested replacing the power supply caps (as did the german site), claiming that fixes some issues. Any input on this?

As far as the battery goes, I know the unit shown at the top of the thread has a 28 yr old battery in it, which needs replacing. Is the procedure similar to that of a 3478A? Do the same rules regarding grounding applicable? The service manual says to leave it powered up, rather than adding a backup source for the battery alone.

PaulAm · « **Reply #9 on:** February 25, 2015, 07:45:55 pm »

I'd check the 180V supply to the display; it's also possible there might be a problem on the display driver board, it looks like the display is multiplexed. Shouldn't be too hard to track down, it looks like they're a little current starved. I think K04BB has the service manual with schematics.

Someone referred to the display as a VFD; it is not. This is a high voltage gas discharge display (Panaplex). Essentially it's a7 segment nixie array.

Dang, now I want to get a 1071

dom0 · « **Reply #10 on:** February 25, 2015, 07:48:57 pm »

Oh, well then, the CRT tricks won't work on a gas discharge display I guess.

In the worst case one has to build a replacement display from e.g. LED displays. Shouldn't be too hard with the schematics, and even worst case (taking multiplexed input from the 180 V? drivers) is something any MCU can do. It'll look like crap, tough.

gilbenl · « **Reply #11 on:** February 26, 2015, 04:20:22 am »

Here's a link to the manual/schematics that I have. Any identifiable places of interest?

https://www.dropbox.com/s/z8sgw2ffkddjjgj/Datron_1061_1061a_1062_Service_Manual_complete_with_schematics_and_partlist.pdf?dl=0

lowimpedance · « **Reply #12 on:** February 26, 2015, 04:38:07 am »

So did you check for excessive ripple from the power supplies ??.
Since these displays have pins and usually plug into a socket why don't you try one of the suspect ones in the good unit, that way you will confirm if its actually the display that's faulty.

gilbenl · « **Reply #13 on:** February 27, 2015, 12:50:53 am »

Quote from: lowimpedance on February 26, 2015, 04:38:07 am

So did you check for excessive ripple from the power supplies ??.
Since these displays have pins and usually plug into a socket why don't you try one of the suspect ones in the good unit, that way you will confirm if its actually the display that's faulty.

Truth of the matter is I've been distracted and haven't investigated much, but now that I get a sense that the fix may be a sourceable part...

First good inspection of the bottom:

Same view from the top-Must've dismissed as dust.

Maybe some ripple.

dacman · « **Reply #14 on:** February 28, 2015, 10:07:23 pm »

You have an LS101, which means it has a D50 connector in the back and no ratio function. (It also means that it has option 12 AC and Ohms and no current function due to option 12 ACV taking up both the option 10 ACV slot and the current slot. Option 12 is twice as accurate as option 10.)

The battery does not look like OEM. It looks like a replacement battery that I've used. I'd guess that it would test at 3.6 Vdc (i.e., like new).

In the picture with both DMMs, are they supposed to read the same? If so, is the Front/Rear switch in the Front position?

An IC that is soldered in (other than the CANs) is the RMS IC on the bottom. It's the one the black heat sink is clipped onto. It's on stilts

gilbenl · « **Reply #15 on:** February 28, 2015, 11:48:53 pm »

Quote from: dacman on February 28, 2015, 10:07:23 pm

You have an LS101, which means it has a D50 connector in the back and no ratio function. (It also means that it has option 12 AC and Ohms and no current function due to option 12 ACV taking up both the option 10 ACV slot and the current slot. Option 12 is twice as accurate as option 10.)

The battery does not look like OEM. It looks like a replacement battery that I've used. I'd guess that it would test at 3.6 Vdc (i.e., like new).

In the picture with both DMMs, are they supposed to read the same? If so, is the Front/Rear switch in the Front position?

An IC that is soldered in (other than the CANs) is the RMS IC on the bottom. It's the one the black heat sink is clipped onto. It's on stilts

Thanks for the info-Work with these units or just hear this info along the way? The top and bottom are not supposed to read the same-I took the picture while testing a HP6920B Meter Calibrator that I just restored. The top is measuring the output, bottom is measuring the reference zener. The battery was replaced 4/28/95 per the sticker on the back and does read 3.6V. Now that I have it open, I'll probably replace it anyway. While I have in unplugged, is there any reason why I can't stick 3.6V across the terminals (with the battery still connected)?

Disassembled the back panel today and removed the LM309K at the epicenter of the mess. The craftsmanship on this thing is remarkable. Once the screws were out of the regulator, the leads just about about fell off.

And the culprit:

I recently fixed an electronic load and couldn't help using this as an excuse to give it a shot. The Fluke and no-name on the bottom show input current and voltage, respectively. The 3478A and 1062 show ouput current and voltage, respectively. Input was ~14.2V from the E3610. As suspected, the best it could muster while maintaining 5V was about .55-.6A. Beyond that, the output went erratic. Max output current was about 1.2A. Ordered a new one today at $37+sh and paid $10 for the spec'd wire. I'm optimistic that this will leave me with 2 sweet DMMs.

dr.diesel · « **Reply #16 on:** March 01, 2015, 12:01:59 am »

That is a beautiful display, very nice unit my friend.

dacman · « **Reply #17 on:** March 01, 2015, 02:03:12 am »

I've worked on 1062s for many years. It seems that acid flux was used to solder in the LM309K (on them all) and it was not cleaned off.

On your other (working) unit, check the 10uF/350V (if OEM) capacitor on the back panel. I've replaced that cap in maybe 90% of the units I've worked on. The display can look fine even if this cap is shot.

Also on your other unit, check the six blue (if OEM) electrolytic capacitors up front on the DCV board. I've replaced some or all of them in possibly every unit I've ever tested them in.

For the battery, I've replaced a few, and there is a recommendation to replace them every five years, but I don't remember actually testing a bad one (in the 1062), even if it looks OEM. It seems that the (CMOS) circuit draws so little current (Nano Amps), that it's the shelf life of the battery that matters most.

If you try to power the battery circuit, I'd use a current limiting (and isolation) resistor. The circuit draws Nano-Amps. I personally would not worry about it. I sometimes clear and reprogram the constants anyway.

dom0 · « **Reply #18 on:** March 01, 2015, 08:44:20 am »

Those batteries can live for quite some time. I put "replace RAM battery in 3478A" on my todo lists months ago only to find that is still measures 3.0 V (it's a 3 V cell, not a 3.6 V one), after 28 years. If it doesn't leak and measures fine I'd not bother with replacing it.

gilbenl · « **Reply #19 on:** March 01, 2015, 05:37:08 pm »

Quote from: dacman on March 01, 2015, 02:03:12 am

I've worked on 1062s for many years. It seems that acid flux was used to solder in the LM309K (on them all) and it was not cleaned off.

On your other (working) unit, check the 10uF/350V (if OEM) capacitor on the back panel. I've replaced that cap in maybe 90% of the units I've worked on. The display can look fine even if this cap is shot.

Also on your other unit, check the six blue (if OEM) electrolytic capacitors up front on the DCV board. I've replaced some or all of them in possibly every unit I've ever tested them in.

For the battery, I've replaced a few, and there is a recommendation to replace them every five years, but I don't remember actually testing a bad one (in the 1062), even if it looks OEM. It seems that the (CMOS) circuit draws so little current (Nano Amps), that it's the shelf life of the battery that matters most.

If you try to power the battery circuit, I'd use a current limiting (and isolation) resistor. The circuit draws Nano-Amps. I personally would not worry about it. I sometimes clear and reprogram the constants anyway.

I suppose I'll leave the battery alone for now. They do read 3.6V, although I understand that the voltage output is virtually irrelevant until just before EOL. While I wait for the 309, I'll do some cap testing with the signal gen. Really need to build or buy an ESR meter...

If anybody is interested, I attached a link to this project's album on the first post. A few more high-res images along with those already in the thread.

@dacman-Do you have any experience with interfacing the 1062 with GPIB? Does it play well with agilent-compatible solutions? There are some arduino based GPIB-USB adapters that should give me the output of my 3478 and it would be nice if the same would work with my 1062s. Also, do you know the name/style of the external input socket?

lowimpedance · « **Reply #20 on:** March 02, 2015, 12:00:54 am »

Nice equipment stack there, like the two 'stacked' handhelds

.
Heh that regulator was just hanging on there.

Doing a refurb on a 4920 with 'tired electro's' at work so these old timers usually require a recap.

dacman · « **Reply #21 on:** March 03, 2015, 02:11:53 am »

I use the 1062 with NI cards. I know of no IEEE issues. It seems to work just fine.

The rear input on the 1062 LS series uses a DDM-50P connector. A DDM-50S would mate to it. The pins and sockets of these connectors have a gold finish and they are rated at 1000 VAC or more, depending on the manufacturer. (A generic D-SUB 50 connector would fit but I wouldn't recommend one.)

The round connectors were made by Pye Connectors in the UK. They seem to no longer exist.

gilbenl · « **Reply #22 on:** March 03, 2015, 10:14:07 pm »

Much appreciated Dacman.

I figured those round connectors were unobtanium-I'll likely just swap those out for a current Ampehnol or Neutrik connector.

When I asked about the rear inputs, I was referring to the round connectors that my OTHER 1062 (non LS101) has. It wasn't until you mentioned the 50 pin that I realized the LS101 lacks the round, but has the 50 pin. Nevertheless, the manual appears to indicate that rear inputs of any type have a max rating of 250V (see attached). Could you confirm?

dom0 · « **Reply #23 on:** March 03, 2015, 10:23:45 pm »

Seems like a pure connector limit to me (D-sub pin spacing is too narrow for more than 350 Vpk)...

artag · « **Reply #24 on:** March 04, 2015, 12:33:33 am »

I think the connectors are Pye M7P or M10P (plugs), M7S or M10S (sockets). I have one or two around, and they were mentioned on here somewhere recently too.

I have a 1051 with a matching scanner, and I think another 1051 for parts. That also has a panatron display but I don't know if it has enough digits. Shorter displays were common in pinball machines and a few spares are around, but they've often been replaced with LED displays that replace the driver board the panatron is mounted on - e.g http://www.pinballcenter.eu/catalog/pinscore-pinball-led-display-for-williams-system-p-2149.html?language=english

gilbenl · « **Reply #25 on:** March 04, 2015, 03:20:10 am »

Quote from: artag on March 04, 2015, 12:33:33 am

I think the connectors are Pye M7P or M10P (plugs), M7S or M10S (sockets). I have one or two around, and they were mentioned on here somewhere recently too.

I have a 1051 with a matching scanner, and I think another 1051 for parts. That also has a panatron display but I don't know if it has enough digits. Shorter displays were common in pinball machines and a few spares are around, but they've often been replaced with LED displays that replace the driver board the panatron is mounted on - e.g http://www.pinballcenter.eu/catalog/pinscore-pinball-led-display-for-williams-system-p-2149.html?language=english

Interesting prospect. In theory, there's no reason a kaput gas discharge couldn't be replaced with a micro and LCD. They pretty much elephant walk you through the data structure in the manual attached previously.

So says the guy who wouldn't know where to start... Armchair quarterbacking at its finest! I hope the regulator and the 10uF cap fix the issue.

By the way, tested the caps and the 10uF/350V cap on the PS was shot. The rest, including those up front appeared healthy on the scope. Will post some pics of the traces if anyone is interested.

lowimpedance · « **Reply #26 on:** March 04, 2015, 05:10:56 am »

Which circular connector were you referring to ?. The shrouded ones for the external trig, ratio and rear sig. ?.

I also replied in this old thread in relation to those seen on an 1072.
https://www.eevblog.com/forum/reviews/what-do-you-want-to-see/

And indicated the only current manufacturer of similar connectors. (Positronics)

These were made by various companies in the past, and I know this because I have a drawer full of them in various pin configs and more than one company name on them.
(Amphenol being one)

I remember in the long distant days as an apprentice seeing a similar effect on a panaplex display from a modular style panel meter, which you guessed it was a dud HV electrolytic.
So good chance all will be good.

dacman · « **Reply #27 on:** March 05, 2015, 03:09:15 am »

A manual for an LS version has the following in it (paraphrased, since it's from memory):

---------------------------------------------------------------------
The LS meter adds the following to the standard model 1062:

Rack Mounting Hardware
Rack Slides Hardware
Enhanced True RMS AC (Voltage)
Rear Panel 1 kV Input utilizing a Cannon type DDM50P connector (mating connector DDM50S not included)

Note: Do not switch inputs with more than 250 VAC applied
---------------------------------------------------------------------

Input pins are not adjacent on the D50 input connector (except for the guards which are adjacent). They are on even numbered slots in the middle row of pins. Only 10 of the 50 pins are wired.

gilbenl · « **Reply #28 on:** March 13, 2015, 07:03:59 pm »

Well, I've replaced the LM309K (M1) and HV cap (C6) and unfortunately was unable to improve the display. I have not replaced the 5 or 6 caps on the leading edge of the analog board, but I do have them. However, I don't think they have anything to do with the display anyway. Any other suspects?

A couple lessons learned the hard way for anyone planning to work on these units: I found it extremely difficult to work on the "box" pcb assembly in situ, particularly with regard to the regulator M1. It would've been far less stress on the PCB to just take the whole (literally) thing apart and work on the box assembly alone. Second, there's what appears to be flux residue on quite a few surfaces. Definitely not sloppy and it appears as though it was cleaned following assembly, but over the years the residual has shown and appears to be the culprit that damaged the regulator leads. I took some IPA to the visible areas.

I wish this had worked, but I'm more than open to further suggestions.

dacman · « **Reply #29 on:** March 13, 2015, 09:44:37 pm »

Look on the bright side. If used under IEEE, the display isn't as important.

If I remove the back panel (two Pozidriv screws on each end plus the two in the heat-sink for the LM309K), I usually remove the IEEE board also (and first). Then only cable connectors are attaching the back panel, and once detached, it can be set aside. To get to C6 on the power supply (10 uF, usually), I will also remove the digital PCB because I like to use a Hakko 808 to remove the capacitor.

I think that unit was left on 24 hours a day for years by the user. I don’t think the display can be fixed (unless it is replaced). (I saw a 1062 at eBay for $95 that had a working display, but otherwise needed fixed.)

I used to buy the display, but can no longer find new ones
8.5 Digit Display
Datron P/N: 800017
Dale P/N: PD-09D064-1
Catalog P/N: PD009D06401G02

The Dale P/N is no longer listed on their website
http://www.vishay.com/displays/plasma/

dom0 · « **Reply #30 on:** March 13, 2015, 10:03:47 pm »

There's always the option to re-engineer a display using other technology (e.g. LCD or LEDs)

Rog520 · « **Reply #31 on:** March 18, 2015, 05:31:05 am »

Can anyone tell me what the options (numbers) were on this meter? I'm looking at a 1062 "MT-5" and can't find much descriptive info on it other than a service manual and a very minimal spec sheet. What kind of math functions can it perform?

gilbenl · « **Reply #32 on:** March 18, 2015, 03:12:59 pm »

That one is MT-5 and Opt 80. I've attached a list below. Note that current measurement is an option and a hard one to find. MT-5 doesn't have it (look for the current fuse on the top right of the rear panel). Dacman can probably enlighten us to what MT-5 entails.

AFAIK the math functions are the same for all 1061A/1062:
Measurement Offset (A-B)
Pct Deviation (A-B/C)
Min/Max/Range w/Storage
Upper and Lower Limits w/alert

With option 40, ratio measurements are possible. The only thing that I miss is averaging, which is only available on the 1071/1081.

dacman · « **Reply #33 on:** March 20, 2015, 11:49:30 pm »

A 1062 comes with DCV, True RMS ACV, Ohms, IEEE-488, External Trigger Input, Rear Input, dB, and Computations. (DC/AC Current, Ratio Input, and Analog Output are options.) I’m not that familiar with the MT-5, which may change the base unit, but one had a D-SUB 50 Rear Input and the External Trigger Input was a BNC Jack.

dacman · « **Reply #34 on:** April 06, 2015, 03:35:17 am »

MT5 means that the 1062 can interpret CIIL commands and is compatible with MATE applications. (CIIL = Control Interface Intermediate Language.) Also, Datron changed the dB and Ratio keys (along with rearranging other keys) to Rear 2-Wire & 4-Wire. This means that Local/Remote Guard and 2-Wire/4-Wire Ohms are programmable for the rear terminals (but not the front terminals). (On other 1062s, rear terminal Local/Remote Guard and 2-Wire/4-Wire Ohms need to be externally wired.)

gilbenl · « **Reply #35 on:** April 18, 2015, 08:08:03 pm »

I've been using the 1062s for the last week to monitor some voltage standards I picked up. After about a week, one unit reports a min/max diff of 300uV, whereas the other reports no more than 5ppm. I took the first one out of the setup and attempted to zero the input-Error 4. Switching to manual mode, I found that the 10V ranges and higher would zero without error. This is not the same unit that I've worked on in this thread, by the way.

Today, I began the servicing process. Initial visual inspection showed that C7 on the power supply is blown out, with a hole coming out the side. I checked the 5V and 15V points and they were all fine, albeit a bit noisy. I have to recalibrate the thing anyway, so I pressed on through the check steps without changing anything too much. I'd like to continue to use it to monitor the deviation on the voltage references until a new cap arrives, so I attempted to recal the Lin, Ib and the ZERO for 0.1 and 1V DC ranges. In all instances, Error 4 is the outcome.

Although I don't have a suitable axial replacement handy, I lashed up a couple in parallel and jumpered them in across the bad cap. The bad cap was still in circuit. No improvement on the noise, auto zero or cal zero-The latter two still resulting in Error 4. Finally, I switched over to the ohms range and tried to Autozero and failed on the 10ohm range

Now is where I start scratching my head-I switched to the rear inputs, shorted Hi and Lo and was able to auto zero on the 1VDC range. The Error 4 persisted on the 10Ohm range. When I switched back to the front panel inputs, I was also able to zero the 1VDC range but not the 0.1VDC. The noise remained and was equal on both inputs.

Summary-

1) Input noise on all ranges
2) Auto zero Error 4 on 0.1 VDC, 1 VDC, and 10Ohm ranges.
-Able to auto zero on 1VDC after performing autozero using rear inputs.
3) Zero, Ib, Lin Error 4 on 0.1VDC and 1VDC ranges. Didn't attempt with ohms ranges or rear input.

Any thoughts would be appreciated. The full service manual should be linked on the first page.

dacman · « **Reply #36 on:** April 19, 2015, 03:26:00 am »

The pictures of the (original) unit look uncannily familiar. It looks like a unit I retired awhile back. (I told one of my supervisors about it and was told that our company does not sell anything on eBay.) The writing on the ACV shield looks like my writing and I'd love to see what is under the shield (it's held by three #1 Pozidriv screws).

For your present problem with the other meter, I would suggest getting a decent LCR meter. I use an ESI 253 (which is now serviced by Teagam). The value of C7 is above the ability of the 253, and for it I (sometimes) switch to a GenRad 1689M (and use it in series mode). The 1689M is presently serviced by IET Labs.

Your problem is not unusual, but there could be more than one issue. First, the power supplies need to be fixed (and I call the 40 Volt inverter a power supply). The six (usually) blue aluminum electrolytic capacitors up front on the DC board are for the 40 Volt inverter. They can cause every problem you are describing. I check these (in-circuit) with an ESI 253.

Second, the relays may need to be repaired or replaced. (They can be tested to determine if they are OK.)

Third, there could be a problem with IC contacts. A mere reseating may not work. I use DeoxIT (and a commercial cotton swab) on the contacts.

I would also suggest checking C6 in the power supply. I cannot get kelvin clips directly on the left side of it, so I use an extension on the left side (if looking from the front). The extension I use is made of a test lead grabber with a solid tinned copper extension sticking out the top that I soldered onto it. (If I use an alligator clip and banana lead for the extension, I do not get good readings.) I also use an ESI 253 for C6. The (in-circuit) readings I get with the ESI 253 are completely reliable for all these capacitors.

gilbenl · « **Reply #37 on:** April 19, 2015, 05:21:22 am »

Luckily, I purchased extras of the 6 caps up front when I replaced them on the other. Will swap them out tomorrow. C6/7 need to be ordered.

The ESI 253 is a really nice meter. I should invest in a quality LCR. I've stalked an HP 4261a on ebay for a while, too.

Here are the pictures...PM me if its familiar.

gilbenl · « **Reply #38 on:** April 19, 2015, 06:01:37 pm »

Murphy.

I replaced C45-C50 this morning and was able to input zero the 0.1VDC range. The noise persisted with modest improvement. I was unable to make it through the Lin and Ib steps of the calibration. So, I went to the analog assembly troubleshooting procedures to see if that yielded any leads. Everything was checking out until I got to step 28 (see image). Almost immediately after touching R20, I see smoke and R184 is burned. And, it just so happens that I don't have any 27ohm resistors to replace it with.

I'm assuming that the resistor failure was unrelated to the test being performed. It appears to be directly connected to C50 (10uF, 63V cap) so maybe the resistor was the initial failure point and couldn't stomach the correct output from the new caps(?)...

Just going to put this thing aside until I can fix the power supply properly and replace the resistor.

dacman · « **Reply #39 on:** April 20, 2015, 11:44:19 pm »

If you are using the on-line manual, there is a mistake in the hardware alignment section (more than one actually).

For step 18 in the analog section (above the sequence that you posted), the test point should be TP19 vice TP13. From TL8, TP19 is on the other side of R109.

For step 19, TP33 seems a little blurry to me in the on-line manual. It's between C54 and R206.

gilbenl · « **Reply #40 on:** May 26, 2015, 12:11:36 am »

Shortly after my last post, I found out what caused the smoke. You can even see the left over flux. Dumb. Walked away for a while.

A month later, I've gotten down to repairing the power supply. Found this little 5ohm resistor stuck on the output of the voltage regulator. Not shown in the schematics and wasn't present in the other 1062. Anyone know of a good reason for it to be there?

If you're considering one of these meters, the voltage regulator (LM309K), C13/C15 on the power supply and C45-C50 on the analog board have needed to be replaced on both 1062s. About $50 to digikey thus far (mostly the 309). Picked up a 309 on ebay for $10. We'll see how it fares.

dacman · « **Reply #41 on:** May 31, 2015, 12:38:41 am »

Quote from: gilbenl on May 26, 2015, 12:11:36 am

Found this little 5ohm resistor stuck on the output of the voltage regulator. Not shown in the schematics and wasn't present in the other 1062. Anyone know of a good reason for it to be there?

A similar series resistor is in some units. Its presence would be dependent on the individual LM309K. The output of the LM309K, from a datasheet, can be 4.7 to 5.3 Vdc. In the 1062, there is an adjustment that is supposed to be adjusted to 5.1 Vdc at test points on the digital PCB. The output of the LM309K can be adjusted up, but not down. The markings on the resistor tell me that it is 0.15 Ohm, with R representing the decimal mark. I would guess that the 5V regulator needs to supply about 500 mA, in which case the resistor would knock the output down by about 0.075 Vdc.

gilbenl · « **Reply #42 on:** June 03, 2015, 11:03:33 pm »

Thanks dacman. When measured, it read 5ohm. I believe the markings (learned something new), but will re-measure for psb fault point.

gilbenl · « **Reply #43 on:** June 24, 2015, 01:55:29 am »

This is getting a bit annoying. Got around to replacing the following: C6, C7, C13, C5, M1 (see attached diagram), and the burned resistor on the analog board. Turned the unit on and W1 sparks and dies. I know these rectifiers can go bad, but somethings not right. Any ideas?

gilbenl · « **Reply #44 on:** August 23, 2015, 02:00:50 am »

Shes ALIVE! (second from top)...and calibrated on the 1V and 10V ranges. Replaced the W004 and the associated tantalums and reseated the IC's on the analog board. Cleaned off all the sticker gunk from both. Look fresh off the line in 1987:box:

Vgkid · « **Reply #45 on:** August 23, 2015, 07:26:30 am »

I think your 3456 is not well.

gilbenl · « **Reply #46 on:** August 24, 2015, 06:42:54 pm »

The 3456 is very well albeit slightly out of cal. The error, however, is known and constant. Its used primarily for ratio measurements and math functions (variance). At some point, I may crack the panel and fiddle with the pots, but its the aversion to messing with a potentiometer-based calibration that has kept my fingers out.

To that point, this is why the 735A has never been adjusted-I just get someone to measure it on a cal'd DMM every so often and note the value. The pots in the unit have been untouched for a lonnnnnng time and are (arguably) less likely to budge with vibration etc. Not to say the 735A is anywhere near the same instrument, but its for this reason that the Fluke 732B uses BCD switches instead of pots.

HP735A-

Fluke 732B-

dacman · « **Reply #47 on:** October 11, 2015, 04:05:03 am »

The 15 V power supplies are critical in these Datron's. One lazy trick I use to check the rectifier is to measure the input voltages to the (variable) 15 V regulators. With 120 Vac of line power input voltage, the unregulated dc voltage is usually around 22 Vdc into the regulators (both + & -). If one is 22 V and the other is below 20 V, I'd replace the rectifier. (With a new rectifier, the unregulated input is normally 24 V.)

This is actually based on the specifications of the instrument. The input of a passive voltage regulator normally requires 2.1 V more than its regulated output in order to work reliably. If the Datron is rated for 115 Vac ± 10 % line power input, then at 103.5 Vac, there needs to be 17.1 Vdc or more into the 15 Vdc regulators. At 115 Vac, there should be 17.1/0.9 or 19 Vdc into the regulators (minimum). At 120 Vac, about 20 Vdc. (There also needs to be 7.1 Vdc minimum into the 5 V regulator with 103.5 Vac line power, which is over 8.2 Vdc with 120 Vac line power.) If a value is below the minimum, the Datron would not pass a line regulation check.

gilbenl · « **Reply #48 on:** October 21, 2015, 07:47:11 pm »

Quote from: dacman on October 11, 2015, 04:05:03 am

The 15 V power supplies are critical in these Datron's. One lazy trick I use to check the rectifier is to measure the input voltages to the (variable) 15 V regulators. With 120 Vac of line power input voltage, the unregulated dc voltage is usually around 22 Vdc into the regulators (both + & -). If one is 22 V and the other is below 20 V, I'd replace the rectifier. (With a new rectifier, the unregulated input is normally 24 V.)

This is actually based on the specifications of the instrument. The input of a passive voltage regulator normally requires 2.1 V more than its regulated output in order to work reliably. If the Datron is rated for 115 Vac ± 10 % line power input, then at 103.5 Vac, there needs to be 17.1 Vdc or more into the 15 Vdc regulators. At 115 Vac, there should be 17.1/0.9 or 19 Vdc into the regulators (minimum). At 120 Vac, about 20 Vdc. (There also needs to be 7.1 Vdc minimum into the 5 V regulator with 103.5 Vac line power, which is over 8.2 Vdc with 120 Vac line power.) If a value is below the minimum, the Datron would not pass a line regulation check.

I really appreciate the explanation. After replacing some caps and a few resistors on the power supply I powered on the meter and saw a flash out of the back of one of the rectifiers about 2 minutes later. So she was certainly cooked. I ordered the parts, but walked away from the project due to block exams and the boards sat on a shelf above my bench for 3 weeks. When I got back to it, I ultimately rebuilt the 15V power supply and got great voltages. I put the boards back in only to find that the digital board had issues. There's no display except when "test" is depressed. It fails to progress beyond the first relay click and the panel in unresponsive.

Long story short, what I ultimately did was combine the analog, digital and AC boards from LS101 with the bad display and this 1062. The frankenmeter that came out has the LS101 option (enhanced AC) plus the ratio input, rear input and external trigger of this 1062. I think this was the best solution all around.

Of note, I did happen find that the GPIB boards were not interchangeable between the two 1062s-I assume due to FW differences. I recently picked up a 1071 (which is AWESOME), but lacks the GPIB option. I started another thread regarding adding the GPIB option to the 1071. M18 is unpopulated on my unit and there's a pin from M29 labeled M18 enable. Not sure if either have anything to do with the GPIB option. There's very little mention in the manual about the GPIB option and how its enabled. If anyone has any info please share.

dacman · « **Reply #49 on:** October 22, 2015, 03:42:43 am »

gilbenl, you need to use an electrostatic mat and wrist strap.

gilbenl · « **Reply #50 on:** October 22, 2015, 05:26:24 pm »

Quote from: dacman on October 22, 2015, 03:42:43 am

gilbenl, you need to use an electrostatic mat and wrist strap.

This is true. Sometimes in life, however, you find yourself buying the bicycle before the helmet.


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: Datron 1062 (Read 33635 times)

Share me