HP 1741A oscilloscope - restoration project

[MuchMore]

Hey 


I am new to this Forum, so I guess a short introduction is necessary:


I am:
- a mechanical engineering student
- 22 years old
- from Austria (not Australia )
- (have) some experience in electronics
- quiet frustrated with my first oscilloscope


Now to my Problem:
I recently bought my first oscilloscope an HP 1740. After some testing I figured the time/sweep assembly must have some bad switch/contacts. I carefully removed the main sweep board and immediately saw that there was someone in before me. After cleaning the contacts I spotted that one "flyleg" was bend so I brought it back to the standard position and recognized the metal was very weak, after a closer "look" it snapped off......


does anyone have a good idea how to fix "my" mistake?? replacing the leg seams quite difficult. And I guess HP does not sell replacement parts!?!


 
P.s.: All remaining grammar mistakes are placed on purpose, to ensure an pleasing reading experience...

[tekfan]

Hi! Welcome to the forum.

Probably your best bet on replacing the switch wiper is to find a similar sheet of metal and make the part yourself.
The sheet metal is very thin so don't expect your local hardware store to have it in stock. It also must be quite springy so it doesn't deform when you turn the switch.

You can ask in the hardware stores if they know where you can get spring steel sheet (''federblech'').
You can also try to find a mechanical clock because the spring inside the clock is just the right material. It may be too thick but still worth a try. If you have a mechanical egg timer it probably has this type of spring inside. The only thing you need to do is to cut and bend it into the correct shape. You could also try to find a watchmaker or jeweler and as if he's got some similar metal.

To disassemble the switch to get the wiper contacts out you will have to do the following:
On the other side of the switch there are several marks where the two halves of the switch were sealed together by melting the plastic. You will have to carefully remove this plastic dimples so that the two switch halves can be separated. You can even try to drill out the plastic. Just be careful not to damage the rest of the switch.
Once you have the switch apart you can remove the broken wiper contact and another one so that you know the exact shape of it.

To put the two halves together you will probably have to use some sort of glue. It's easier to remove afterwards if it needs repair again.

Here you can see the 4 melted plastic dimples that hold the two halves together:

[HackedFridgeMagnet]

Yes very light spring steel.
I did one with a banana plug and an angle grinder, It worked but would have been no good, long term.
I then got some spring steel and was set up to fix it when somebody gave me another scope which I then cannibalised for parts.

[MuchMore]

Hey


thanks for the fast replies.
After some thinking I remembered that there was a useless Battery Tester in my drawer.
And sure enough there where two flylegs inside.
The idea of removing the molten nipples is risky, since the plastic is soooo old I fear it will shatter while Drilling. 
I came up with the idea of removing a pocked, with a dremel, on the front big enough to fit the new flyleg and glue it with epoxy.
However if I mess this up it is over...


Since the new flyleg is at least coated with copper is it a longtherm issue off rupping on the goldpats?
I could get a Nickel-solution to make it like the other once..??


 


P.s.: some rant from me: I HATE THE IMPERIAL SYSTEM, now I can go to an expensive hardware store to get does *** Hex keys that will fit the screws inside the knops. 

[vk6zgo]

The problem with the Imperial System is all the little inconsistencies,like the fact that US Gallons are smaller than British/Australian ones,so that a "44 Gallon drum" in Oz translates to a "55 Gallon drum" in the USA.
They are both,thankfully,near as dammit to 200litres.

Old Brit type equipment is likely to have Whitworth,UNC,UNF,BA (AAARRRGH!),or BSF (double AAARRRGH!) screws,while the smaller  screw sizes used in US equipment are of a incomprehensible & completely incompatible series.

In the larger sizes,UNF is pretty much common to both,some UNC,SAE,& Whitworth are compatible in very large sizes.

Even metric screws have weird things,though.--- small Japanese metric screws were different to the same nominal size German screws---though this may have been sorted out in more recent times.

[SeanB]

Metric at least has defined threads and pitches, and taps and dies are available for them all in a single series ( ok it is a 20 page pdf) for ease of use.

[Kilroy]

Since the new flyleg is at least coated with copper is it a longtherm issue off rupping on the goldpats?
I could get a Nickel-solution to make it like the other once..??

The biggest thing to watch out for is that surfaces of anything coming in contact with the gold is a totally smooth, polished finish. The slightest bit of surface roughness will tear the gold right up. Just because it feels smooth to the hands does not mean it is so. Check everything under at least a 12X loupe to be sure...smooth stuff can look particularly horrible under magnification. I use very fine Japanese waterstones for polishing wiper surfaces for these types of repairs. You can *not* use fine wet/dry sandpaper....3000 grit wet/dry will leave a much worse surface than a 3000 grit Japanese waterstone finish. Always carefully compare the finish on the original contact surface with the finish on any potential replacement material before you consolidate the repair. Any impatience or shortcuts in the procedure will reward you with a ruined switch surface.

The HP 1740A/1741A machines are beautiful instruments, perform well beyond their specifications, nice to work on when/if the time comes, and well worth the trouble to repair them when they do require a bit of attention. I would plough ahead with the repair if everything else looks good with the scope. The experience alone counts for something.

Incidently, since the board requiring the repair is the main sweep pcb, I would check the Sprague 109D 100uf 30v wet tantalum capacitor when your are in there. After much use this cap had a tendancy to fail, which kills the slower sweep speeds. It's usually not particularly catastrophic when it happens, but since the sweep board will be out anyway I would just go ahead and replace it. If it is anything other than a Sprague 109D then it has probably already been replaced....hopefully with an equivilent milspec substitute.

[MuchMore]

@Kilroy
    Hey

    mhmm polishing the pad is a very good idea... however I was a Little bit fast. I already put in the new flyleg and glued it back together. (was a real pain in the ass job) 
    I don't know if it is very practical to polish them when they are inside the Frame, or if it would break another one, and there is no way to separate them now (the epoxy is really tough)
    I have to think about that.

    About the cap: you mean C11 ?? I can't read the name on the part. If you mean C11 than yes it look really crusty, but the slow timebases work just fine (for now).
    I don't have the exact replacement part, however I can write it on a list for the next Mouser order.

    So it seams you know which parts are up for an replacement. It would be amazing if you can write a small list containing parts that are prone to fail?


 
for now the scope works, but the knops for the vertical Settings are very "funky", means: you turn them and the line jumps around like crazy, but it is stable when you leave it alone. Is there an easy way to clean the pots, or do I have to replace those? (I have not jet removed the main vertical board)
and the second Problem is, when I Change the Volts/div the ground Level jumps around. Is this an calibration issue or an Hardware failure??

 
I really appreciate your help,    (and I like the positive Response to the metric System)

[lowimpedance]

 Neat refit job on the switch there.
Sounds like you need to clean the vertical switches as well.

 Couple of things to check are the power components on the power supply (including the connectors for any fatigued joints and re flow any suspect looking ones. Also make sure you lubricate the mechanical surfaces of the timebase switches don't need much but will help with the "feel" of the switch detents when rotating. One other failure I have had was a power resistor in the vertical amp hybrid 'bias' go high in value killing all vertical signal getting to the vertical deflection stage.

[Kilroy]

Looks like you used a wiper off something else for the repair so you should be fine, no worries there, and you saved futzing about fabricating a special part.

C11 is the right one. Often fails open when it goes and you don't get much, if any, warning that it is near death. Suddenly your slower sweep speed is a flickering flat trace originating from somewhere mid screen.  This cap is specified at 20v which I suspect is probably towards the low end of the requirement...possibly to keep leakage and ESR down as low as possible, since this tends to go up for equivilent capacitors having higher voltage ratings.

You can replace the original Sprague 109D wet slug with a good solid tantalum if you prefer. I like the Kemet T140 as a substitute...not inexpensive but very low ESR, low DC leakage, good heat performance and stability. I usually go up a bit in the voltage rating but these HP 1740A/1741A machines are very solid, and would go for years and years with no trouble on the originally specified components so if you want to stick with the original 109D 20v cap there's no reason not to.

Sounds like you need to go over all the controls and give them a good clean and flush with isopropyl.

[MuchMore]

Ok I guess that I am not willing to spend 1/8 the money I paid for the scope on an cap the may fail in the future... and when it fails I will replace it with a good cap that will do the Job.

@lowimpedance I can't find an power-resistor on the vertical-amp?? which one do you mean??

The high voltage supply look not sooo good, can someone take a look and tell if there is a part that may be at the end of its lifetime??

[lowimpedance]

That would be R3 on PCB A5 (see circuit diagram) fed from 43V rail. (HP1740)

The High voltage PCB has a little 'soot' on it which is normal for circuits running at high tension (Old CRT TV's for example), so nothing from your photo's looks too bad actually.
 Maybe a gentle wipe with a soft cloth (and a little IPA) to remove the soot.
Check the soldering for any suspect joints, one last inspection and put the cover back on , and don't worry about component failure until it does!!.
Clean your vertical attenuators.

[tggzzz]

I've recently restored a 1740 that was more-or-less functional before I bought it; it now works well. Nice machine.

The remaining "suboptimality" is that the digits on the timebase knob/dial are hardly visible, partly due to grime on the transparent plastic, partly because the charaters are worn away. I'd like to clean the transparent plastic and replace the legend. How do I mechanically disassemble the knob/dial itself, so that I can access the legend?

(To state the bleedin' obvious, I can get the dial out of the machine).

Thanks.

[tautech]

A 1740 manual should be available from one of the links in the first thread in the Repair board.
HP used retaining screws that require very fine hex keys to remove the control knobs, you will need the correct sizes. (imperial)
Search other restoration threads as some senior members have unbelievable skills and tricks to do exactly what you need.

I've also got a 1740 for a wet day project, had another a few years back before I got into DSO's.

[tggzzz]

A 1740 manual should be available from one of the links in the first thread in the Repair board.
HP used retaining screws that require very fine hex keys to remove the control knobs, you will need the correct sizes. (imperial)
Search other restoration threads as some senior members have unbelievable skills and tricks to do exactly what you need.

It came with the repair manual, I'm pleased to say, but it is silent on disassembling the knob.

I had a quick look a while ago, and couldn't be sure it was a hex screw; I wondered whether it was a foobar, where "foobar" is effectively an expanding coiled spring shaped like a very thick danish pastry . Now, if only I could remember the real name for a "foobar".

Either way, I'll have another look, and thanks for your reply.

[SeanB]

Roll pin........

[tautech]

A 1740 manual should be available from one of the links in the first thread in the Repair board.
HP used retaining screws that require very fine hex keys to remove the control knobs, you will need the correct sizes. (imperial)
Search other restoration threads as some senior members have unbelievable skills and tricks to do exactly what you need.

It came with the repair manual, I'm pleased to say, but it is silent on disassembling the knob.

I had a quick look a while ago, and couldn't be sure it was a hex screw; I wondered whether it was a foobar, where "foobar" is effectively an expanding coiled spring shaped like a very thick danish pastry . Now, if only I could remember the real name for a "foobar".

Either way, I'll have another look, and thanks for your reply.

Just checked, they are definitely hex grub screws that hold the timebase knobs on.
2 sizes, 1/16 and 0.050.
the 1/16 is easy to acquire, the 50 thou not so easy.
Both sizes came in a cheap Fuller set #130-8030 metric & imperial to 10mm/3/8 that I have no hesitation to recommend for workshop/lab use.

[tggzzz]

Roll pin........

<sigh>Thanks!</sigh>

[coppice]

I think the 1741A was the only really satisfactory analogue storage scope ever made. Their only big problem was premature failures of the tube. As long as it failed in under 90 days it was a warranty replacement, so HP paid the bill. If the tube failed out of warranty you threw the scope away. The tube was most of the cost of the scope.

[David Hess]

I think the 1741A was the only really satisfactory analogue storage scope ever made.

Why do you say this?  I really like some of the Tektronix analog storage oscilloscopes with the notable exceptions of the 464/466 and operating any of their variable persistence tubes in bistable storage mode.

[coppice]

I think the 1741A was the only really satisfactory analogue storage scope ever made.

Why do you say this?  I really like some of the Tektronix analog storage oscilloscopes with the notable exceptions of the 464/466 and operating any of their variable persistence tubes in bistable storage mode.

The Tek scopes had a number of problems, but the bleed alone was enough to make them pretty awful. I always thought that was rather odd, as they had the bleed totally controlled on their graphics terminals. You could turn one of those off on Friday, and on Monday morning the same 4096x3072 resolution image was still there in its full glory.

[David Hess]

Why do you say this?  I really like some of the Tektronix analog storage oscilloscopes with the notable exceptions of the 464/466 and operating any of their variable persistence tubes in bistable storage mode.

The Tek scopes had a number of problems, but the bleed alone was enough to make them pretty awful. I always thought that was rather odd, as they had the bleed totally controlled on their graphics terminals. You could turn one of those off on Friday, and on Monday morning the same 4096x3072 resolution image was still there in its full glory.

I have never had a problem with this on the ones I have used but maybe I do not understand the issue.  I find the stored image retention on my 7834 rather annoying actually.

Maybe this is an aspect of their storage transfer CRTs which allow very fast single shot writing rates?

[tggzzz]

Just checked, they are definitely hex grub screws that hold the timebase knobs on.
2 sizes, 1/16 and 0.050.
the 1/16 is easy to acquire, the 50 thou not so easy.

The hex keys have just arrived and the 050 fits nicely. Thanks.

But the grub screw is frozen in position sufficiently tightly that I don't want to break the key or strip the head

I've doused the screw in WD40, and will see if that has any effect. Then it will be a light machine oil or penetrating oil.

If that doesn't work, are there any standard techniques I could use?

[SeanB]

Try tapping gently on the hex key to shock the screw in the shaft, and if that does not work take a soldering iron and heat up the hex key and thus the grub screw to loosen the oxides that are holding it, or loosen the locking compound. Heat it till it just starts to smell, then try to turn it loose, and if no turning try to tighten then loosen. Then hit it with freezer spray to thermally shock it and it should come loose.

[David Hess]

I've doused the screw in WD40, and will see if that has any effect. Then it will be a light machine oil or penetrating oil.

If that doesn't work, are there any standard techniques I could use?

My ordered list of methods to remove set screws is:

1. Applying light machine oil and waiting.
2. Using a soldering iron to heat the screw.
3. Sacrificing the screw and replacing it once it is out.

I have a special 1/16 hex key where I filled down the flats on the end to make it into a hollow ground 0.050 hex key so it fits very tightly and will not strip the screw.