EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: Joopha on December 04, 2021, 08:53:06 pm
-
Today I received my "new" tester HP3490A that I bought on eBay for 50euros. It start up but measure always 0 in all modes and ranges. I already verified all supplies are correct and the input signal enter the frontend module but then I measure always 0V at output of dc amplifier. Does anyone have some idea?
-
Hello.
Did he come from Sicily?
-
yes it did.
-
I would think you probably have to first isolate which part of the circuit is faulty. The DC amplifier output feeds the A/D converter. You could try
injecting a voltage (current-limited) at that junction and see if you get anything out of the A/D converter as shown on the 3490A's display. You could have faults in the front-end switching, attenuators, DC amplifier, A/D, or logic/display circuitry. Try to narrow it down first. Then get out the schematic, inject a voltage on the front terminals, and follow your way through the signal path until things look bad.
-
Hi,
I verified the signal to be measured doesn't reach the DC amplifier because the solid state switches are not working.
The inputs of the Eeprom that drives the solid state switches are always low so I suspected any problem in logic circuit at timing level.
I verified the main clock signal doesn't reach the main counter because the divider by 2 that provides the clock has the clear bit low (always clear) and this signal (LCDC) comes from a 4 to 16 line decoder. Looking at the decoder it seems all inputs switches between 0 and 5V but all outs are mostly at 5V except for pin 7 that switches, so I'm suspecting about this IC. It is the HP 1820-0702 but my feeling is that it is a standard SN74154N that is obsolete. Do you have any advice?
-
HP 1820-0702 is listed as a Fairchild 93L11 in the HP cross-reference list. At first glance, it looks like a 74LS154 should also work, although I didn't read the datasheets that carefully. Both chips seem to be available on eBay, although I would stay away from any Chinese seller.
There are also some digital section self-tests and flowcharts in the manuals that might be worth trying.
I don't envy your task. The 3490A is an algorithmic state machine and you have to plow through it. Mine, fortunately, was an easy fix - power supply problem.
-
Hi Guys,
Good news: my HP3490A is alive again!
Eventually the fault was due to the U23 (1820-0586) a 6 channels logic inverter. I discovered the issue because I saw pin 8 of U23 was low that is not possible because pin 9 is low as well so I expect pin 8 high as out of the logic inverter. I checked all the channels of U23 and most of them were not working.
I spend some time to find the replacement but at the end I fond it on an old HP board.
After the replacement I checked the DC measure on a 9V battery and it seems working fine.
Now The real work is to calibrate the tester. I ordered a AD4540 voltage reference and I can borrow from a friend of mine a HP 8.5 digits multimeter.
I never did a calibration so any advice is welcome.
-
There are lists online to find out some of the equivalent parts for the HP custom numbered parts, of course some were custom made for HP and have no commercial equivalent part.
1820-0586 is a 74L04N.
David
-
Basically I fixed the multimeter because I found the original part.
Now What I need is to calibrate the multimeter. As I wrote I only have a precise voltage reference and a calibrated 8.5 digits multimeter.
Any advice about how to calibrate the multimeter as accurate as it's possible with this pour staff?
-
To calibrate the 3490A properly, you need an adjustable DC reference, an adjustable AC reference, and a set of precise resistors. You didn't say whether or not your "precise voltage reference" was adjustable. The calibration procedure is shown in section 5 of the service manual.
For instance, for DC calibration you would need 1 mV and 0.1, 1, 10, 100 and 1000 volts. If you don't have that, you might get away with cheating by using a stable power supply capable of those voltages and your 8-1/2 digit DMM to verify the voltages. In other words, say you are calibrating the 100 volt range and you set your power supply near that value. The 8-1/2 digit DMM reads 99.586723 volts. Then adjust the 3490A so it displays that number.
Similarly, you might get away with a function generator for low AC volts, but good luck on the higher ranges!
For ohms, you could use resistors out of the junk box and similarly adjust the 3490A to read what the 8-1/2 digit DMM does. But you will have to be careful on the higher ranges as temperature coefficients of the resistors plus you touching them with your hands can alter the readings when you switch from the DMM to the 3490A.
The manual says that, at least for resistance calibration, you can use resistors that are up to 10% off from the nominal decade values (1K, 10K, 100K, etc.). But for DC and AC, I'm not sure what inaccuracies would eventually result if, for instance, you used 40 VAC to calibrate the 100 VAC range.
-
Hi wn1fju, thanks for your detailed list of warnings. I'll try to setup a procedure to do a fare calibration and post the results so we can review.