I would like to thank the group. My HP 8594E is now working correctly. I basically cleaned the attenuator inside contacts with deoxit then checked the attenuator by itself on a bench using a simple bread-board setup to actuate each solenoids. After it tested good, I put it back into the analyzer, ran the calibration and it passed without errors!

I have two easy questions:
1: I still have 1 long slim philips screw left. Can’t remember where it goes. Anyone knows?

2: the current CR2477 battery reads 3.0v. What’s the threshold where it can’t keep the ram from loosing its data?

The 7448 is "likely" to work fine, given the circuit you showed in the first post.
Install a socket, and try it. Worst case, you have to plug something else in.

Pete

Hi Greg,
Thanks for the datasheets and Jameco information.? From your mention of the 9307 using active pull-up, I searched for more details on this.?

While looking thru the Jameco 7447/7448 Family data sheets, I noticed the 7448 is shown as Active High and its description actually mentions internal 2Kohm pullup resistors to drive a buffer transistor for the display segment.

I measured out of circuit resistance on the old 9307 chip between segment output pins to VCC pin and it was around 1.6K for all segment pins.? I'm assuming this would be a rough measure of the 9307's internal pull up for comparison.

Here's a screenshot of the Jameco datasheet showing the 7447 & 7448 segment output.

Since the 74LS48's are available, thinking about ordering a couple of these and maybe try them on the breadboard first.? Would there be any interface concerns with using a 74LS48 vs the 7448 with the other older TTL chips in the counter?

Thanks again!

DB,

?

My last message was rather haphazardly written and didn’t include as much information as I had intended.? It was after midnight and I was trying to get some sleep before a 5:00 rise.

?

Looking at the schematic you provided it is obvious that any part substitution will either require a device with an active pullup or you will have to provide same with external resistors if it has open collector outputs.?

?

I looked at the 7447/74LS47 and the pinout is again the same as the 9307 but has open collector outputs same as the 74LS 247 previously mentioned.

?

From all of this I suspect that you have a few options to consider as well as other decoder/driers out there that may also serve the purpose and are still readily available.

?

Greg

An EEPROM dump would be a great asset, saving much grief in people that lost the 54720 mainframe calibration due to a stinkin' dead battery. The 54717 "calibrator" hardware is just a brain dead pair of coax cables and a limiter (which can be omitted, anyway).

I looked for one a couple years ago with no success. They exist in printed form but it appears noone has scanned one yet.

Hi again

On electric/electronic tests:

I've changed three 2N2222 transistors on A6 rectifier board, one mising (due to transformer hit I presume), one leg broken and one corroded. Diode drops, fuse continuity and caps leakage test done too.
Due to lack of information about real current consumption and since my plan is to convert (and in some way, redesign) the power supply, I managed to do the same thing I did on my 8663A restoration a year ago: fantastic and colored 1.5 mm2 (16 awg) wiring to feed the rails using power supplies and transformers. For better and safer current monitoring conections, I opted to use banana 4mm plug set.

I fed the 5 volt rail (with 11.8V) yesterday, and with 2.3A drawing, the logic circuitry looks working fine (error display, knobs working, keys responding). It's a first step, I know. But the 19 amps I estimated for this rail seeing the fuses ratings is far away. After this first attempt, I thought how wrong my estimation was: how could it draw 19 amps with five logic boards (plus plugin)?if the biggest and way more populated 8663a draws 10-12 amps in that rail?. Doing the power conversion,?it makes more sense to think that the current in 8350B is about 5.4 amps @ 5 volts, which makes more sense seeing the actual hardware installed. I never stop learning ^^

The voltage regulation is made using LM317 and LM337 mostly, so the input voltage could be a lot higher than outputs, that reduces the secondary wire gauge on the transformer for developing the same power. That's the reason, I think, for the monster heatsink used in the 8350B, which has the size and weight of a brick.

Still thinking how to adapt the 8566B transformer to this unit, with one less secondary and only one center-tapped. A bit challenging, but I think it is possible.

Regards,
Ruben

My unit has a ton dust and some white and green powder coming out, as you can see in my photos. Nasty but no fatal, only surfaces afected. So, after all, this little fella was lucky.


Tony,?definitely your unit was exposed longer to the corrosive elements. Luckily my unit doesn't seem to have contacts or wires corroded. I sand the edge conector on HP-IB and A7 regulator board pcb,?but they had any severe corrosion, just tarn a bit. Gold plating doing a good job there. I think this particular unit suffers a leaky Ni-Cd battery pack in the past, someone replace it (but no clean the unit) and the major corrosion stops there, because current battery pack looks leaky but not too much (anyway I get rid of it, and think on use a Ni-Mh, like I did on my 8663A).

Hi Ruben,

Looks like you are doing a great job of bringing it back to life. I can sympathise because I went through more or less the same process a few months ago. I think my 8350B might have been stored at the same place as yours :-) It had bad motherboard corrosion - especially in the edge connectors. I treated it with white spirit vinegar and then lots of distilled water followed by a gentle "bake" in the fan oven at about 45.C with the door slightly open to completely dry it out. That caused domestic strife!.? The biggest problem was the power and data connectors to the plugin as the ribbon cables had corroded through and many of the wires were floating about. I ended up having to make completely new ribbon cables. I've added a couple of pics - they look like yours!

Good luck and looking forward to seeing your progress on getting it working.

regards

Tony

Hi Runin,

Deoxit D5 works extremely well on all electronics, pcb with parts, edge connectors etc. the can is small
and a bit pricy.

I restored a HP-8663a with a 2 cans from the inside out after ?sticky fluid damage by others before purchase. Nothing has ever returned after 12 yearsI check every 12 months.

hood luck with the resto.

Regards

Gerald
VK3GJM

Hi!

Some updates and photos for those who are interested:

- On the cleaning: finally?I disassembled everything because corrosion?reached many gaps and I found it necessary to sand and brush each piece to get rid of that white powder crap. Nylon brush for major parts and teethbrush for pcb, fine sandpaper for aluminium structural parts, warm water and soft dish soap. And then some sunlight to evaporate and dry. Some pics of dirty/cleaned parts. Time consuming task but very satisfactory. Some of them was very dirty and corroded, as you can see in the photos. I brush and sand even the sockets and it's nuts, because rear panel was one the most affected by corrosion.

Next post on electric/electronic tests.

Regards,
Ruben

A rather rare one.? Crosses to a industry part number 9307PC (plastic package) or 9307DM (ceramic package).? Is currently stocked by Rochester Electronics but they want you to purchase a quantity of them for over $300.? I see that there is a single 9307DM on ePay if that is all that you need but is a bit pricey.

?

?

Data sheet:

?

?

Without digging further something like a 74LS 247 might work if the outputs meet the needs of the display.? It is pin-for-pin physically compatible with the 9307. ??Jameco Electronics appears to have them for a buck and some change.? And if you enter “decoder driver” in their search window it will bring up several other components but you would have to compare data sheets to see if they would be suitable for the application with respect to pinout & electrical specifications.

?

Greg

开云体育

Has anyone had any joy repairing on of these.

Dave

Is this supposedly a "stock" HP diode or do you suspect it has been repaired B4 with something link a 1n34 or 1n914?

Also, is there any kind of DC block in the system (that disk could be a capacitor. If no cap, why cutoff at 100K? If they are saying that because of chopped DC, then a parallel capacitor would fix that.

Any diode detector without a DC block will be acceptable.

Cheers!

Bruce

Quoting Radu Bogdan Dicher <vondicher@...>:

See?/g/HP-Agilent-Keysight-equipment/album?id=272489

My 8471A, diode is fried, (well, I did find it in a dumpster...) working on replacement, will update results when done...

Dave Wright <davewrightsplace@...> updated the album 8471A Internals: Disassembly of 8471A RF detector prior to diode replacement.