Alex,
Those Vishay resistors from China are the real deal.? They are pulls and affordable.? That is if they are from the company I have purchased them from previously.? ?A year ago I purchased qty.75? ?10K and 40K S102C resistors from a Chinese source and they we original Vishay.? If you want to purchase state side look to Digi-Key.? The? S102 series is the way to go.? Resistors you are looking at with 10ppm/C are not much better than the Dale resistor I designed into the 3468/78 meters.




Your comment:???the service manual seems to be very critical about which mode to select for calibration and recommends to to do so in the ohms mode that the instrument will be used (...the most). So it sounds like one would always have to make a small compromise, and having one of each basically sounded like being able to eliminate that compromise. Or not really.

My current comments supersede the manual for I wrote those sections of the manual -? LOL

Theoretically the offset between 2-W and 4-W cals is as I pointed out in my last post is 41mOhm when shorting the inputs but more realistically 60-90mOhm with the presence of thermal offsets and contact resistances when ZERO caling the instruments with the shorts applied to the inputs.? It makes no sense to cal in 2-W ohm with a short at the input.? You want to cal in 2-W mode with your test leads attached and shorted at the probe tips.

You will not find a source for the RED Devil resistors for they were never marketed to the industry.? They are proprietary components that were made at HP Components in Loveland Colorado for use as attenuator and gain setting resistors in the HP instruments.? This was before Vishay came along with a better product.? The RED Devils are precision wire wound resistors with low TC characteristics and tolerances.? You will find them in all the early high performance multimeters from the 60s and 70s.? Even the 3456A has some of them.


ANSWER:
And on the other hand there are no 190M ohm resistors available at all, the closest standard value seems to be 187M ohm. Closest possible exact match for a series configuration would be to use four 47.5M ohm, but they are 100ppm rated, so not sure it would be useful long term beyond just finding out if the ranges would fall in compliance after connecting this resistor combination in parallel with R201.

The value is not critical for testing as long as the shunt resistor in somewhere near 200 Meg:? 175 - 225 Meg would do the trick.

I will try to get the first installment of the HP3468A/78A design series out this weekend.
George

Comparison readings between 3478A and 3468A
-----------------------------------------------------------------

Used my little resistor standard box I build around a resistor array 1k - 10M 0.1% and a couple of extra separate 0.1% resistors for the 10 and 100 ohm positions. All readings taken in 4W ohms mode and with the chinese kelvin cables, at 28 C ambient temp.

Resistor?????? 3478A??? ??????? 3468A
--------------------------------------------------
10 ohm??? ?? ? 10.013x??? ???? 010.030
100 ohm??? ?? 100.037??? ??? 100.028-9
1.0001K??? ??? 0.99997K????? 0.99959K
10.01K??? ????? 10.0099K????? 10.0070K
101.01K??? ??? 101.010K????? 100.968K
1M??? ???????????? 1.00093-4M? 1.00065-6M?
10M??? ?? ? ? ? ? 9.97xxM??? ???? 09.99xxM

The "x" denotes digit(s) with unstable continuously fluctuating reading.

Results from the Red Devil Resistors
-------------------------------------------------
20K?????????????? 19.9721K?????? 19.9662K
98.975K??????? 99.022K???????? 98.980k

Far from perfect, but not that bad.
Wish it would be as easy to get a stable multi-voltage standard to be able to check all those VDC ranges as well.

I am looking for information to trouble shoot scope that just quit?

there is no help or drawing in service manual?

开云体育

开云体育

Sent from Samsung by banging a nail into a piece of wood.

Nigel,
I will see what I can do.? The same connector blocks were used on the HP3421A portable data acquisition unit so look out for those.? Unfortunately when you purchase cards, the connector blocks and housing are missing.? Best to look for completed units with cards stuffed into them that have the external cables cut at the blocks.

There is a chance I will be in the UK within the next 20 days on business.? If that goes thru and I find spares I will bring them along.? Will be flying into Heathrow.

Some people earlier were looking for floppy drives for the 9122/9133 series of disc drives.? If there is a need in the UK, I could drag some of those along with me.? ?Let me know and we can work out a price for the units.

George

Hi George,

Very nice explanations. I will follow your advice and calibrate the 3478 as well on 4W. Its just that I was a bit less trusting about its calibration than the 3468, so my reasoning went to use 4W on what I perceived as being the better of the two. But as you point out the '78 having the the lower ohms range, it definitively would make sense to use it for low value 4W measurements. Although I have to say that with the zero cal on the 3468 and the chinese kelvin cable I was for the first time able to read and verify the value of some 0.01 resistors I had purchased some time ago to use as shunts in a power supply project but never used them. They are more like a U shaped piece of wire, and so far all of them. And same goes for basically any other low value resistor in my parts stock that I threw at the 68, it measured up quite well within the expected results.

I really need to compare resistor readings between the 68 and 78, which I did once, but then also this time write down the values for comparison and post them here just for the record. I will recalibrate the 78 in 4W and do a comparison of readings.

My reasoning to have one DVM on 2W and the other on 4W stems from the fact that the service manual seems to be very critical about which mode to select for calibration and recommends to to do so in the ohms mode that the instrument will be used (...the most). So it sounds like one would always have to make a small compromise, and having one of each basically sounded like being able to eliminate that compromise. Or not really.

For the time being and following your suggestion, I will not attempt to calibrate the 30k/300k ohms ranges by using the Red Devils. No need for absolute accuracy.
In regards to your suggestion about R201 did some research at Mouser, my preferred supplier:

To substitute R201 outright there are no 40.0K series S102 resistors available for that exact value. Closest I could find is Vishay PTF series 40K 1/8W 0.05% 10PPM/C at $3.38 each, 58 currently in stock. Would that be still an improvement over the existing resistor?

https://www.mouser.com/ProductDetail/Vishay-Dale/PTF5640K000AYEK?qs=sGAEpiMZZMsPqMdJzcrNwo3vE%252BHgilrQScfmxiejoUM%3D

Second best options would be another Vishay PTF series 40K 1/2W 0.1% 10ppm/C. One does cost $4.99 and the other only $0.89. Frankly, with a side by side comparison, can't seem to find any differences between these? two other than the selling price (both prices are for qty one unit purchase).

71-PTF5640K000BYEK
https://www.mouser.com/ProductDetail/?qs=41svs21nsHBzvOZxn1wkhQ%3D%3D

71-PTF5640K000BYEB


And there is actually a Chinese seller on ebay offering used Vishay S102K at $5, 40K 0.01% which as pictured are obvious pulls. But given ebay's history with remarked fake parts mainly from China I doubt this could be trusted to be the real deal, and even if it where given aging and accumulated drift, etc would probably preclude this as an option in this case. If this where the only option, figure probably better off staying with my current R201.
https://www.ebay.com/itm/122674485353

And on the other hand there are no 190M ohm resistors available at all, the closest standard value seems to be 187M ohm. Closest possible exact match for a series configuration would be to use four 47.5M ohm, but they are 100ppm rated, so not sure it would be useful long term beyond just finding out if the ranges would fall in compliance after connecting this resistor combination in parallel with R201.

As many others, looking forward to your video, and when the time comes hoping for some more of those Red Devil resistors that you might want to sell for use as standards. BTW did a G search on them and closest results is for Ohmite "Brown Devil" resistors, a whole different beast. Basically zip on the Red ones.

Thanks for all the Hi-Q information.
Alex

For HP70K, ?HP7000 users:

There is also a thread for HP70K technical & repair issues, ?on the EEVBlog with photos, schematics & interleaved text.



?

Seconded! I have two, and love 'em. They are wonderful units, damn
handy. Nice work George and team.

-Dave

--
Dave McGuire, AK4HZ
New Kensington, PA

开云体育

Hello,

now I am awake again.

IMHO; the next test we must do is to check if the CPU card is farting out a video waveform. If you see the "marching ants pattern" on the LEDs of the card, it is very likely that at least the CPU is alive.

Tam

After power on,all 8 leds were on for less than one second,then 6,then 4,then all leds were off (completed power on self test)
after a little while the leds flashed one after one in sequence.(each time only one led was on).

I kept front panel faced up and powered it on for couple minutes,then?powered it off and on again
while holding one button,for couple times,but nothing changed.

Jack

-- 
With best regards
Tam HANNA 

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

开云体育

Sent from Samsung by banging a nail into a piece of wood.

开云体育

Are you sure it is the YTO?

Newly this error appears.
Does somebody has a reliable source for spare YIG.
In may vna is a 5086-7718 mounted.
Armin
HB9MFL

Tim,
Tell me what it is you are looking for and I can help you.? I was the analog design engineer on the HP3488A.
I designed the 444XX series of I/O cards and have a large supply of all the parts for the mainframe and 444XX series of cards.
That project began in 1981 and completed in 1983.? It was one of the highest selling products the HP Loveland Division ever produced.? ?Tons of them out there these days for very little money.??

I see there is interest in the power supply.? It was not an OEM supply.? The power supply was designed in-house by one of the other engineers on the team by the name of Tom Rodine.? There were 6 of us all together on this project:
Project manager
Analog Design I/O and main frame interface
Firmware / software design
Digital design
Mechanical design
Power supply design

George? Hnatiuk? ?April 19, 2019? 8:32am

开云体育

George,

开云体育

Sent from Samsung by banging a nail into a piece of wood.

开云体育

Sent from Samsung by banging a nail into a piece of wood.