开云体育

This has been a fascinating thread.? Recently watched video on his HP 419A LED upgrade and just had to see more of the history behind the neon switcher.? The technique is both mundane and novel at the same time.? It has certainly given repair folk a hard time over the years with ageing neons and CdS cells.

?

One rather interesting CdS cell type that I came across was in a round metal can with a glass lid about an inch (25mm) in diameter.? They were used in some dirt cheap two component day-night switches for outdoor lighting.? The high current CdS cell would control the drive of a heated bimetallic snap-action switch that was able to switch about 1000W at 240V AC. The interesting calibration technique they used gave me a chuckle.? The face of the CdS cell was painted with some aluminium paint and the amount of light that could enter the cell was adjusted by manually scratching the paint to open about 2-5% of the surface to achieve the required sensitivity.? The hysteresis may have been adjusted by bending the stop for the snap action lever that was driven by the bimetallic strip.? The unit was so simple I first thought they were fake but testing proved they worked pretty well.? Looking at the scratches on the batch of 20 it was clear that three different people were involved in calibrating these units because you could recognise the marks like signatures.? A bit like modern laser trimming of ceramic hybrid resistors.

Kalle
--
Kalle Pihlajasaari
Lahti, Finland

Hi John
Sorry but 45409 is not a UK post code

If still available, I would like the ones for the 8901B and the 8640B.? Please let me know if they are available and how much to send to Zip 35601.

Thank you,

Harold Foster

Hello,

how much would this:

5. 8753C User's Guide and Quick Reference

be to Hungary?

Tam

With best regards
Tam HANNA

Enjoy electronics? Join 15k7 other followers by visiting the Crazy Electronics Lab at

I have the following manuals available for the cost of shipping from
45409. Most are originals though a couple are good quality copies, and
most are in HP binders.

1. 3585B Operating, Service Vol. 1, and Service Vol. 2 (3 binders)

2. 8712ET/ES and 8714ET/ES Service Guide

3. 8920A, 8920B User's Guide Rev. B and
HP8920B HPIB Programmer's Guide with IBASIC (2 binders)

4. 8901B Modulation Analyzer Operation and Service

5. 8753C User's Guide and Quick Reference

6. 3561A Dynamic Signal Analyzer Operating Manual (probably a copy)

7. 8640B Signal Generator User and Service (original bound version)

8. 8902A Service Manual volumes 2, 3, and 4 (3 binders)

Thanks,
John

I will study your writeup with great pleasure once you have it organized and published.

Remind me how you are matching your H11F's and canceling the remaining offset.

Dave Wise
________________________________________
From: [email protected] <[email protected]> on behalf of Sergey Kubushyn via Groups.Io <ksi@...>
Sent: Monday, March 02, 2020 10:38 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP 419A Null Voltmeter - Purchase

OK, I've put everything together finally for the first run as a whole and it
worked like a charm. Not fully assembled yet (covers not installed) but
otherwise pretty complete.

There are still screwups so the main board (that big one replacing the
existing amplifier) will have to be re-spun

I would like to buy these!
Please contact me directly at fourdesigns@...

Hi,

This is pretty similar to what the 8640B signal generator does. You
can lock the oscillator to the quartz time base via the counter. Then
by changing the counter timebase you can fine tune the frequency. One
caveat is that the counter will not indicate the change as the
generator is locked to the counter. And, probably doesn't have enough
resolution anyway. So to find the 'fine tuned' frequency, one has to
use an external counter.

Szabolcs

DW <wilson2115@...> ezt írta (id?pont: 2020. márc. 1., V, 22:19):

At 2020-03-02 03:51 PM, Mark Litwack wrote:

Page 3 of the *CONFIG* menu for the sync settings, not the CAL menu.

Thank you! I would never have figured that one out on my own. Now I can actually see the full screen on my HP8595E. The collective knowledge of this group continues to amaze me.

Steve Hendrix

Hi Steve,

On 2/29/20 3:53 PM, Steve - Home wrote:

I’ve done the “make program disk” and “load program disk” and it works a treat.

Thanks for the confirmation! I also had good hopes it would work. Now
all I need is an image to load ;-) But that requires someone that can
get LIF files off of a disk.

My machine does have both PAL chips so it can support both flavours of
the OS.

Best regards,

Job

I also converted an 8510X to 8510C. Someone mentioned awhile back that
you couldn’t take a program disk from one serial numbered machine and do
a “load program” onto another 8510 as the programs contained serial
number data. That doesn’t appear to be the case.

The 8530 to 8510 may be a different thing, though, as there’s a PROM or EPROM specific to the 8530 on the security key board. I don’t have an 8530 so I can’t give it a try.

Steve
WB0DBS

Thanks for the tips Vladan, did not consider that type of impact from
vibration. I definitely will check out the vibration when I move out the
machine to get to the rear ports. I also hate the weight of the thing ;-)

The 1800RPM is a good option then as it is yet another 9dB less noisy.

Best regards,
Job

Check the rectifier feeding C5. The capacitor will get hot if there is AC on it from a shorted or leaky diode.

Robert G8RPI.

If you haven't seen yet Mr. Carlon's Lab just did a teardown and restore with custom LED's and driver:?https://www.youtube.com/watch?v=vrmwql2msbU&t=639s

On Mon, 27 Jan 2020, Dave Wise wrote:

OK, I've put everything together finally for the first run as a whole and it
worked like a charm. Not fully assembled yet (covers not installed) but
otherwise pretty complete.

There are still screwups so the main board (that big one replacing the
existing amplifier) will have to be re-spun -- 1V and 1mV calibration
trimpots swapped, all trimpots CW/CCW also swapped somehow (it makes it
_LESS_ sensitive when turned CW i.e. moves the needle CCW,) probably try to
find better trimpots instead of those single-turn cermet chinese blue ones.
All other boards are almost fine (RGB LED silk is backwards on the front
LEDs PCB but that is not biggie, won't do anything with that) and work
exactly as expected. Internal wiring changes all tested and worked OK (not
all that extensive but anyway.)

All ranges work OK. 3uV is wandering within one upper scale small division
(half of lower scale division) that is tolerable at such sensitivity and
shitty probes I used for input. Thermal EMF from those probes is significant
at 3uV full scale so better wiring is needed for such sensitive
applications. Chinese probes connected to series 1Mohm/100ohm AB Standard
Resistors (100ohm on far end of the probes) give something like 300nV EMF at
room temperature that is readily measured by HP419A :) Shorting input
terminals with copper bar makes it return to zero that it held steady
without any noticeable drift for 4 hours.

Have a lot of photos from the process but no time to organize those right
now and design a web page. It is not finished yet anyway until that main
board is re-spun that would take me a week or two to re-design, another week
for JLCPCB to make and ship and then some time to assemble everything.

Can put some unsorted unedited photos on my web site as one zip file if
anybody is interested but that would be raw photos and archive would be
huge. Will process everything and write a web page for those when done (that
would also include all schematics, PIC source and hex and so on) but have to
finish everything first.

I wish I had a 419 I could try this on. Quite the magnum opus, Sergey. I
look forward to studying it.

I too played with replacing neon with LED. (In my 740B DC Voltage Standard
/ Differential Voltmeter.) I used Lumiled "Luxeon" series phosphor-white
parts. You can easily get 10x or 100x the illumination that was provided
by the neon lamps. The limiting factor is avoiding reducing the input
impedance too much. I found that no matter what intensity and duty cycle I
used, I could not improve much over the original neon excitation. One
interesting and frustrating observation was that performance improved
greatly if I kept the excitation running 24x7. Leave them dark overnight
and next morning they're near-dead all over again.

Fabrication was finicky because I had to insulate to 500V to meet spec. I
programmed what I had on hand, PIC12F675. I included adjustments for
frequency, duty cycle, and modulator-demodulator delay. The latter is
helpful for capturing maximum efficiency if you use H11F1 in the
demodulator where its offset and TC are don't-care but it switches so fast
it's out of phase with the modulator.

But since I had already replaced the entire modem assembly with a quad of
H11F1, hybrid LED/H11F1 is bench experiments and not actually tried in a
running instrument.

Regards,
Dave Wise

I had so much fun with the programming. The 740B has two separate chopper
amps, running at different frequency, 162Hz and 95Hz. I programmed one
12F675 to excite both. Originally I interrupted every 65us (greatest
common divisor of the two periods) but the ISR ate up almost all of the
processor cycles. I wrote a program for my PC to explore other divisors,
and discovered several that were within 0.1%: 79, 101, 121, and 129. That
gave me breathing room. By the way, anyone wants the program, just ask.
________________________________________
From: [email protected] <[email protected]> on behalf of Sergey Kubushyn via Groups.Io <ksi@...>
Sent: Sunday, January 26, 2020 11:35 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP 419A Null Voltmeter - Purchase

On Sun, 26 Jan 2020, Marco Marchesi wrote:

Oh, this is a rare find :)

I have 2 here and none of those worked properly. I spent a year trying to
get them to work with small quick fixes and that didn't work.

So I bit a bullet and now finishing a full rework using most of the parts
but throwing away and adding quite a lot of new stuff.

First of all I'm keeping photoresistor assembly. It is a very low leakage
part with no chopper signal feed-thru. However I'm removing neons and
replace that neon board with an assembly made from 2 small boards -- one
base/carrier board having exactly same dimensions as original neon board and
two very narrow thin boards replacing neons. Those have 4 LEDs on them, 2 on
each side facing photoresistors. I don't know what photoresistors they used
there -- it might be CdS or CdTe or a mixed one -- and they are sensitive to
different colors. This is why I'm using Yellow (~590nm) and Red (~680nm)
LEDs on each side that cover any of the photoresistor kinds. Neon has 2
major spectral lines, one at 585.2nm, another at 703.2nm so I'm close to
covering both. I do already have boards, experimenting with different LEDs.
That is almost over, waiting for a shipment of final parts (1206 LEDs from
Würth Elektronik) from DigiKey, expected to arrive this Wednesday.

Then, entire back part with power supply and chopper driver is replaced with
one expensive relatively complex board. It carries 2 18650 LiIon batteries,
very low noise switching converter to +/-15V with very low noise linear
regulators providing +/-13.2V for instrument power, isolated low noise Null
Voltage supply (that replaces original mercury battery), and a small PIC
(16F15323) that drives chopper LEDs with proper waveforms and Power LED
which starts flashing when batteries are about to die.

This board also carries full featured batterry charger and off-line flyback
SMPS feeding that charger so batteries can be charged in situ, without
removing them by plugging mains cable and pushing "Quick Charge" button. It
takes 3 hours to fully charge batteries and then they can run for more than
24 hours without re-charging.

That is quite big and expensive board -- it uses several expensive chips
from TI and Analog Devices (actually Linear Technology), couple of
transformers and some other stuff -- but it is well worth it. It is 4-layer
board with lot of attention put to thermal design as the entire charging
part generates lots of heat and some parts run at 80-90 degrees Celsium
under peak load, when battery is fully discharged.

All parts of that board are debugged and tested and I do already have PCBs
and stencil on hand. The only part left is a small thin PCB holding 3 fancy
TVS diodes that acts as heatsink for those -- it is already ordered from
JLCPCB but it is Chinese New Year now so they will only start making panels
and stencil for those starting February 1 and it takes ~5 days from start to
getting them delivered by DHL Express.

Then there is another small board that goes between front panel and
sub-panel. Nothing fancy there, just one RGB LED (RG for charger and B for
Power indicator) and one Orange LED with a small bridge and current source
running off mains voltage to indicate that HP419A is plugged into mains
that it should not be for normal operation. There are also 2 LED lenses
installed into front panel, one into existing "Line" neon hole enlarged to
round .281" for the lens and additional one above "Quick Charge" button. I
do have those boards and already have 2 assembled and installed in my 2
419As.

But that is not all yet :) I do also have a new main board replacing
existing one with proper capacitors and other stuff. This one is essentially
a copy of what was there but with some snafus fixed and better parts. This
one is untested as of right now although I have the PCBs and even have one
board fully assembled.

I expect to finish everything in 3 weeks, maybe a month. I'm documenting
everything as I go making lots of pictures. Once it is done I will release
all schematics and everything else into public access.

As I order PCBs ten at a time I will have 8 full sets of bare PCBs left when
I'm done that I could give out for those brave enough to undertake their own
rebuilds for a nominal charge (just shipping and what I actually paid for
those boards that is not much.)

Stay tuned if interested :)

Hi All,
I'm looking for this unit in WORKING condition.
Is someone willing to sell?

Let me know, Thank you

Cheers

---
*
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
*

Got a new 3.5 Female cal kit today and a thumb drive with the data. ?It’s non HP but the cal data seems not bad ? Not that I really know much. But did get a chart of numbers ?

Moved the .CK and XKT files to floppy. ?Can see them using file utils back can’t figure out how to load them. ?

Any help is appreciated. ?I am hoping not to have to manually key them in. ?

Thanks
Dave

?

JF,

Thanks, I have contacted you, (I think), through your contact in on this forum.

Sorry for the delay on the response. It does not have option 004. I tried what Don B had said about leaving it plugged in. I usually don't. I do have a GPS 10Mhz standard that I leave on 24/7. Even with this plugged in I still get the error on start up. I left the HP3336b in an on all day while I was working in my lab. Near the end of the day it stated to make and odd buzz type noise. It kind of sounded like it was coming from the power supply area. I turned it off. I was checking around it and it seemed that the cap C5 was hot compared to C8. C5 was a 50 deg higher then C8. I used a infra-red temp gun. I pulled the power supply out and removed C5. I tested it on the tester and it passed ok. The ESR was .005. I can't do a leakage test on it. The?capacitor is a 1000uf 35v. I have a 1000uf 50v I think I will try to see if it is the problem. I will let you know what happens.