Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Need help with HP 1222A scope, no sweep

Status
Not open for further replies.

clickfix

New Member
Hi!
This is my first post here, so maby I should introduce my self; I'm from sweden, on my last year studying to become an EE but I've hade electronics as a hobby many years before I started my education.

A couple of days ago I bougt these two nice, old (1977) HP scopes really cheap, one works perfectly, and one is broken. When in sweep mode I get no x movement on either channel, but when in x/y mode i get movement on the x axis, indicating that the x-amplifier works. The x possition adjustment is also off, letting me only move on the right half of the screen.
Luckily I also got the service manual, so I set up the scope and fed it a sine as mentioned and started probing it.
It seems the problem is in trigger mechanism, this consists mostly of TTL-logic, which thankfully is pretty easy to meassure. But I'm having problems with isolating the source of the malfunction, since the circiut is pretty messy with a lot of feedback paths (combination of logic and analog circuitry).

I got a scanned version of the manual, which can be found here: https://tinyurl.com/7slcrrz

I also marked a few testpoints and suspected areas in the schematic of the trigger generator:
**broken link removed**

The testpoints, with reference waveforms from the manual that I tested are marked with green or red depending on working status, the green ones have a waveform, the red ones are flat (DC).

The JK-flipflop U7, marked with pink boxes, was one suspected component, it controlls the nand-gate where the waveforms doesn't get through. But I tried and manage to set it by pulling some of its pins low, so I don't thik It's here.

The strange latch U8, orange box. Doesn't giva an output but it seems to be controlled from U7.

The light blue box contains the chop-oscillator, this is what alternates between A and B to display both at the same time i believe. This is completly dead, no matter how i set the times/div and A/B switches (witch suppusedly changes its speed / turns it off). Could this some how cause the trigger to malfuncion or vice versa? I'm having a hard time figuring that out.
The most promising error source here is the three caps C29, C30 and C47, marked with red circles, C29 / 30 are large tantalum caps (says 1000, = 1000uf or 1000nf??). But I don't have the tools to measure them in circuit (and probably not out either).

I should really appreciate some professional help here, before I completly dissassemble the scope to do more thorough meassurements / try changing components.

I should also point out that all supply voltages are stable and good, no ripple, so no bad caps in the PSU.

Please have patience with any spelling errors since english is not my native language :)
 
U8 looks like a "one shot", it gives a pulse of a specific width when triggered. I can't find part numbers in the manual, can you read them off the chip? You say the red ones are "flat", but you don't post levels, this would help.

Yes, that flip flop in the pink box upper left certainly looks suspect. Need a part number for that so we can see if CLR should be /CLR, and need to know what the level on that is.

You SURE you got +5V on this thing? I notice your active signal chains seem to be the ones driven offboard from analog sections on the +12V line - sometimes this will give you "phantom" activity in logic getting powered through their inputs. On these kinds of chips part of the signal can go through the body-drain diodes and give enough power to get some misleading activity on a chain of gates, like in that Schmitt-trigger section.
 
Last edited:
thank you for your answer, this is the second forum i post to and the first answer i get.

U8 is indeed a one shot; 74122N

As for the measured points (not shure if you can read the numbers from the picture):
11: 3.7v
14: 3.5v
16: 0v
17: 0v
18: 0v
19: 5v
20: 0.5v
21: 0v

U7 is a 74H106PC, the level on CLR is 5v, clear is active low, so the chip should be active.

Yes it does seem strange indeed, like you say missing 5v would explain a lot of the problems I'm experiencing, but when I measured it looked fine 4.98V, totaly smooth waveform. Also the voltages on testpoints 11 and 14 seems to be a bit odd, 3.5v = logic 1, but it should output 5v...

I will measure vcc pins of all the chips tomorrow to see if they are really 5v, maby i missed some thing. Now it's bed time!
 
Nope, all the IC:s in the trigger section are recieving 4.98v, with a noise (the 50hz (Swedish standard) ripple is barely vissible on my digital scope) of 6-8mv.

I am new at troubleshooting and repair, but from what I heard the components that are most likely to give in first are the capacitors, then maby the transistors could be a little bit sensitive, but TTL IC:s ? They are not even CMOS, should be quite robust?

One thing that might also be of interest is that the calibration oscillator (pictured on the page I posted in the lower right corner) Is slightly off, I shoul get 2kHz @ 0.5v squarewave, but it is 2.5kHz @ 720mv, with pretty bad rise/fall time. Interestingly I get almost the same error on the other working scope. The timing caps in this circuit is two large film capacitors of 100nf each.
First I thought that this was not connected to the rest of the circuit in any way but now I see that it goes to U8, the one shot. This does reciev the waveform when the Auto off switch is pulled. But still nothing on the output of U8.
 
from what I heard the components that are most likely to give in first are the capacitors, then maby the transistors could be a little bit sensitive, but TTL IC:s ? They are not even CMOS, should be quite robust?

I'm afraid the part about the caps is largely a myth - antique radios and new Chinese power supplies often have bad electrolytics, this has given rise to the idea that caps are always the first to go, but it just isn't true - I have replaced more power transistors, switches and connectors with bad contacts than anything else. Also just plain old broken wires. Still, if your calibration osc frequency is off by a whopping 25% that sounds like it might be those film caps. You say the rise and fall times are bad? Did you adjust the trimmer in the scope probe you are measuring it with to be sure?

I replaced LOTS of TTL chips in arcade games in the 1980's, I can tell you for a fact they go bad, and I hated having to unsolder them. Seems like it would be a good idea to replace U7 - it should be toggling along, and isn't. Before you remove the chip, you might try clipping pins 15 and 14, right near the board, and bend the pins up and test them without the connections to the NAND gate and the one-shot, because the other possibility is that BOTH those other two parts have blown inputs. That does not seem as likely, though. Probably that 74H106.

Do you have a temperature probe? In the old days I used to find bad TTL by looking for the hot one. This worked great on banks of identical chips, like video shift registers. It was less helpful on a mixed bag like this - but if you found a really hot one, that was usually the problem. You need a probe with a small thermal mass, like a JK thermocouple.
 
Last edited:
Good tips! Even if i don't manage to fix this thing I will still learn a lot along the way, that is worth a bunch. But I will not give up just yet.

The probe is adjusted correctly (checking the cal.osc on the digital scope) the waveform has lag, looks like a lowpass filtered square wave.

Yes, I guess I will have to try and change U7, sadly that chip seems to be rather hard to find, but maby I can find a pin compatible JK latch. After some googeling I did find the chip on some canadian surplus site, but the shipping would probably take a while.

Sadly none of my meters can measure temperature, but I do have a digital steak thermometer :) but it probably won't be accurate enough, I will try touching them and see if I can feel it some of them are hot.

What could have caused a problem like this? Some kind of line voltage spike or abuse? Or does IC:s age as well, mabe they where bad from the begining.

Seems like a cool job repairing arcade machines, near my parents house there is an arcade machine repair shop and as a kid I allways wanted to work there, now they are mostly repairing laser game equipment from what I heard.
 
I could rarely make guesses at the causes of this kind of chip failure. Sometimes it would be something connected to a line leading off the board, these might have been static. The video section was often driven near their speed limits, this increases power dissipation, overheating.

Some of the errors had funny human causes: once we had replaced the "image ROMs" in a game and sent the board back saying we did so. Unfortunately, a dipswitch setting got changed to "cocktail mode", which makes the screen default to an upside-down position. The operator installed the board, saw the screen was upside-down, read that we had replaced the image ROMs, and figured we must have put them in backwards. So he pulls them out, turns them around, plugs them back in, and... BANG!
 
He he, yeah but seems like a logical thing to do if you have no clue of how this stuff works :) What was the "coctail mode" for anyway, was it so that you could install the screen upside down if you wanted to?

So I found a replacement chip that is pin compatible to the 74h106; 7476. It should be easier to source, all the big companies has it, I even found it in the largest store here in sweden, guess what it cost? Allmost 15$ and thats without 30% tax even! That supplier is f**ing joke! Thats almost as much as I payed for the entire scope :p
But I shall call the local shop and see if they have it tomorrow, that is unlikly, so I probably have to buy it from Germany.

I left the scope goning for 30minutes and touched all the chips and all of them felt cool. Non of them are hot lika a transistor shorting out anyways.

It is still a mystery why the chop oscillator is also dead thoug, but while I'm at it I should try change that chip to, It's just some nand gates branded with HP:s strange codings, so I should have something compatible in my parts drawer.

One question about changing the components: do you think I could manage to desolder the chips from the "wrong" side? the internal design of the scope is pretty nice and easy to measure on. But It would be a pain to dissasemble It, lots of wires going to the board, an it is attached to the front panel. If I use solder wick could I get enought solder out so I could remove them do you think or is that bad practise :)
 
You could try to desolder from the top, cut the pins off and do them one at a time maybe - but I think you will probably have to take the board out.

That $15 chip price is something I always explain to these dopes grinding the chips up trying to make money recovering some microscopic amount of gold in them. The chip is worth many times what the gold is worth to somebody trying to repair a vintage piece of equipment.

"Cocktail mode" was for cocktail-table cabinets - galaga-ct.jpg
Same boards, different cabinet. The default mode on some of these was upside-down from normal.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top