scematic for hp3478A?

[0867532]

Hello, i failed cal on my hp3478a. But i failed it in wired way. I just touched cap510(as i remember) with iron and meter told me "uncalibrated". Ye ,ye, they both was floating, i swere. Anyway i find out that i got only 0.3v on sram chip insted vbat(3v). Somewhere voltage drops. Its hard to me trace out ram power,
so has someone scematic? I cant find it.
P.S. ive checked few diods and caps, they looks good. Hope sram is not dead.

[mrflibble]

Hello, i failed cal on my hp3478a. ... so has someone scematic? I cant find it.

Was about to attach my local copy of the service manual, but then I noticed it is still available from agilent: http://cp.literature.agilent.com/litweb/pdf/03478-90008.pdf

[dom0]

Seriously? Did you even google'd "hp 3478a schematic" once? You get like ten links to scans on the first page.

[Guni]

Yes, there is full service manual
http://cp.literature.agilent.com/litweb/pdf/03478-90008.pdf
Better phrase for google is "service manual" then "schematic"

[tautech]

Hello, i failed cal on my hp3478a. But i failed it in wired way. I just touched cap510(as i remember) with iron and meter told me "uncalibrated". Ye ,ye, they both was floating, i swere. Anyway i find out that i got only 0.3v on sram chip insted vbat(3v). Somewhere voltage drops. Its hard to me trace out ram power,
so has someone scematic? I cant find it.
P.S. ive checked few diods and caps, they looks good. Hope sram is not dead.

Welcome to the forum.

Please link your country flag in your profile for local recommendations if needed.
Your request might have been better placed in the "Repair" board where there are many links to manuals in the first thread.
We understand for "newbies" the forum is quite large and growing all the time, spend some time studying where threads should be posted so to keep each of the "boards" more tidy. 

[lowimpedance]

If the sram is dead, or otherwise the memory backup supply has, you will also need to recalibrate.

[0867532]

Yeah, that was realy dum. Ive searched "hp 3478a scematic". Anyway thanks all, who drop manual!
Ive rechecked cr764 and it have >1MR both sides.
Anyway i want to replace it. i afraid it has wired HP part number or so... could it be ok if just replace it with 1N5819(DS:SCHOTTKY BARRIER RECTIFIER)? or with wich i have to replace?

P.S. Ive launced DMM and tested it. No ram failature messages or else, just "UNCALIBRATED". U512 got 4.98V while DMM on. but 0.322V while off.

[dom0]

Uncalibrated means that the calibration RAM contents are invalid. The 3478A uses a checksum for the calibration RAM and can detect this.

Probably COTS diode selected for low leakage and/or low drop. BAT54 could be fine. 1N5819 has huge leakage current (~1 mA). OTOH I'd try a normal 1N4148 first.

[MarkL]

CR764 is an HP 1901-0050.  Keysight's part finder lists it as:

  1901-0050 Diode-Switching VR-80V IF-0.2A IR-0.015uA TRR-2ns DO-35

1N4148 would probably work fine, as dom0 says, but 1N914 or 1N4448 has better Vf/If characteristics to match the above.  Or, Keysight still sells 1901-0050 if you want to pay the ridiculous shipping charges.  There's also someone selling a lot of 12 on ebay right now too.

But the bigger expense is going to be getting the meter properly re-cal'd once you replace the battery and diode.

[0867532]

1N4148 would probably work fine, as dom0 says, but 1N914 or 1N4448 has better Vf/If characteristics to match the above.

Thanks! ill use one of them.(1N4448 or 1N4148 i can get cheaply tomorrow)

Or, Keysight still sells 1901-0050 if you want to pay the ridiculous shipping charges.

I dont think i would even imagine shipping to russia one diod... ahahah

Calibration is problem number 2. First i have to fix it, so meter can save cal.
And i think i found place to cal. it.

P.S. Is there any statistics of % russian members of forum?
P.P.S. but i sill have no idea how i could fail that diod
P.P.P.S. Why HP marks diods "CR" instead of "D"?

[dom0]

CR = crystal (i.e. semiconductor / not a tube) rectifier

[MarkL]

P.P.P.S. Why HP marks diods "CR" instead of "D"?

What dom0 said.

Either D or CR can be used according to IEEE 315-1975 and ANSI Y32.2-1975.

[0867532]

As it turned out cr764 was good.
Scematics says that after cr764 dirrect connected u512 pin 22(suppose to be ok, cos if sram failed on self test it shall give U.C.RAM FAILED or someth like that.)  and it is not shorting circuit. c510(ive desolder it-cap ok) and u515a logic &. Maybe this thing wrong?
but as i get, this logic & doing of "power good" or something like that

[MarkL]

As it turned out cr764 was good.
Scematics says that after cr764 dirrect connected u512 pin 22(suppose to be ok, cos if sram failed on self test it shall give U.C.RAM FAILED or someth like that.)  and it is not shorting circuit. c510(ive desolder it-cap ok) and u515a logic &. Maybe this thing wrong?
but as i get, this logic & doing of "power good" or something like that

When you say you measured U512 as 4.98V, you mean pin 22?  Please be explicit with the measurement information you're providing.

Have you confirmed the original battery is dead?

One thing to try is to get two AA or AAA cells in series with 500ohms or so (in case there's a short) and use them in place of the battery as a test to make sure U512 is being powered (pin 22) when the meter is off.  You can also check for excessive current draw by measuring voltage across the resistor.

And if I understand you correctly, the meter was working fine before you touched C510 with the soldering iron?

[0867532]

You can also check for excessive current draw by measuring voltage across the resistor.

voltage across R765 (470R) is 1.933V so something U512 or C510 or input of U515A in short.
U512 i think is good, because there is no errors while self testing and its not heating up while work from "mains" and gets 4.98V on pin 22.
while "mains" off pin 22 of U512 gets 0.316 V, but battery is 3.1V now

And if I understand you correctly, the meter was working fine before you touched C510 with the soldering iron?

Yeah, but thay both is floating. meter trough step-down transformer(its 120v instead of my 240v.) with no grounding. and iron is just cheap-ass plug in wall. (dirrectly to 240v, no grounding at all)

[0867532]

Ohhh... i findout that i have not same board rev as manual, and there is no c519 and few other resistors. Anyway I find wrong part! CR500 why did they put freacking pointer diod?(or it looks like). So can i put 1N4148 instead?

[MarkL]

CR500 is listed in at least 4 part tables as a 1901-0518, which is this Schottky diode on Keysight find-a-part:

  1901-0518 Diode Schottky Barrier BVR 70V 0.25W THT

But it's also listed in table 5-3 as a 1901-0050 (same as CR764):

  1901-0050 Diode-Switching VR-80V IF-0.2A IR-0.015uA TRR-2ns DO-35

Does CR500 look like CR764?

I can't find a schematic for your version, but changes to the RAM power supply are mentioned in the backdating section.  What is the serial number on your unit?

(I said 1N914 or 1N4448 as a closer replacement for a 1901-0050, not 1N4148.)

[0867532]

So, meter works now correctly. U512 on pin 22 gets 2.8V (Vbat is 3.01V). I could not get  1N914 or 1N4448 today,(but i ordered already few 1N4448) so i put 1N4148 instead 1901-0050(CR764) and 1901-0518(CR500), just to make it up and running.
S\N is 2301A04741 -realy old one
Thanks everyone for help!
Now i witing for diods and going calibration...
P.S.No, CR500 looks like old point clear-glass diod and CR764 looks like normal red and black glass diod
P.P.S.but i cant get why they used different diods(cr764 and cr500)

[MarkL]

So, meter works now correctly. U512 on pin 22 gets 2.8V (Vbat is 3.01V). I could not get  1N914 or 1N4448 today,(but i ordered already few 1N4448) so i put 1N4148 instead 1901-0050(CR764) and 1901-0518(CR500), just to make it up and running.
S\N is 2301A04741 -realy old one
Thanks everyone for help!
Now i witing for diods and going calibration...
P.S.No, CR500 looks like old point clear-glass diod and CR764 looks like normal red and black glass diod
P.P.S.but i cant get why they used different diods(cr764 and cr500)

Schottky diodes have different characteristics.  In this application HP probably wanted a lower voltage drop from the battery to the RAM chip.  I'd have to see the actual schematic to be sure.

You should select a ultra-low reverse leakage Schottky to replace CR500.

[MarkL]

An alternate part number for 1901-0518 is 5082-5509.  A posting over in the HP/Agilent yahoo groups suggests 1N5711 as a replacement:

  https://groups.yahoo.com/neo/groups/hp_agilent_equipment/conversations/topics/31621

1N5711 has more leakage than what I had in mind, but at low voltages it should be ok.  I'd probably still look around a little to see if there's anything else better that's easily obtainable.

[dom0]

So, meter works now correctly. U512 on pin 22 gets 2.8V (Vbat is 3.01V). I could not get  1N914 or 1N4448 today,(but i ordered already few 1N4448) so i put 1N4148 instead 1901-0050(CR764) and 1901-0518(CR500), just to make it up and running.
S\N is 2301A04741 -realy old one
Thanks everyone for help!
Now i witing for diods and going calibration...
P.S.No, CR500 looks like old point clear-glass diod and CR764 looks like normal red and black glass diod
P.P.S.but i cant get why they used different diods(cr764 and cr500)

Schottky diodes have different characteristics.  In this application HP probably wanted a lower voltage drop from the battery to the RAM chip.  I'd have to see the actual schematic to be sure.

You should select a ultra-low reverse leakage Schottky to replace CR500.

Given the very low current consumption of the MOS RAM it probably doesn't matter too much. I think the maximum spec of the MOSTEK RAM in my 3478A is something like 1 µA. Uf of a silicon diode like the 1N4148 is probably in the range of 100-250 mV for currents this small.