HP 1727A scope revisited; vertical amplifier problems.

[Alex_Baker]

Hello all!

I am revisiting this troublesome HP scope after nearly a year off of it, which is not as long as I thought it would be, but being a student without a job I don't have the money to work on any other projects except fiddle with this pile.

If you need any context from the original forum topic here is the link:
https://www.eevblog.com/forum/repair/hp-1727a-oscilloscope-with-multiple-issues/
I will try not to use this topic otherwise.

Starting with updates since I last posted about this, I really have not done anything except run through as much of the adjustment procedure as I could with the equipment I have. This did actually yield good results, I have a nice bright trace and all the intensity/brightness controls work. However I believe that there are problems in the vertical amplifier, symptoms listed below.

Number 1: The distortion at the center horizontal graticule. If I remember correctly this only showed up recently.
[ Specified attachment is not available ]

Number 2: The trace is not centered when a signal is applied, even if the original sweep is centered with no input. changing the coupling does not make a difference.

Number 3: While sweeping the vertical position across its range of motion, the amplitude changes quite significantly.

Number 4: The amplitude seems way off, however it has not been adjusted and I don't want to adjust it until some of the other problems are solved.

I will continue working on this myself, but I thought some of the scope gurus out there would be interested. If anybody needs more information let me know, I will do my best to respond ASAP.

[bob91343]

That seems to be a CRT issue.

[Wallace Gasiewicz]

First; how does that triangle wave look on your other scope?

[Alex_Baker]

The triangle wave looks perfect on my USB scope, the function generator is an older BK precision unit that I just did some repairs and adjustments on for this repair project.

[Wallace Gasiewicz]

I have seen this sort of trace before and it was in regards to an alignment.
I just could not remember if it was an alignment of a scope or function generator.

I believe you are correct in that the problem is probably in the vertical amp.
I have not read the manual for adjustment. But perhaps if you go through the steps carefully you can find the proper adjust point, and if that does not work, the discrete components around the adjust point may be at fault.
I will attempt to get and read the manual

[Wallace Gasiewicz]

Just so we are using the same manual:

https://www.keysight.com/us/en/assets/9018-02796/user-manuals/9018-02796.pdf?success=true

Page 133 / 176  Quote:

"If you have distorted waveform check capacitors for shorts and resistors for opens"

However there are several adjusts that are on this board I would start with the "BAL ADJ"     Mark the position before adjusting it

Look at the schematic (next page) and the board itself. I think these adjustments should be marked.

The problem shown in your trace indicates that the push pull type (maybe my terminology is wrong)  amp in the vertical amp is not starting at "zero" this is U2. (consisting of four transistors) We have to see if it is not getting the correct signal or is mal adjusted.
Half the amp is responsible for the pos part of waveform and the other half for neg.

[Alex_Baker]

I will have to check those things when I get the chance.

I do have the PDF manual, It is kind of a pain to use when you have a paper service manual right in front of you, but unfortunately mine is an early incomplete version.

[Alex_Baker]

I checked the balance, all it does is move the trace up and down, the distortion stays at the center Graticule.

I checked resistors and caps on A5, I don't think there are any open resistors or shorted capacitors, though it is hard to tell since as usual there is a whole host of differences between my scope and the manual.

I did follow the waveforms on the board going into U1 and U2. The signal going into the IC from the delay line looks fine (pins 3 and 6), shown below is the waveforms on U1 pins 4(blue) and 5(red).


This shows the waveform on the output of U1 to U2, pins 12(blue) and 13(red).


Obviously this does not look right, one of the vertical deflection plates is not getting driven properly, any suggestions on things to test.
I am crossing my fingers that U1 or U2 is OK, otherwise this scope is doomed.

[Wallace Gasiewicz]

I think the next thing to do is check the voltages on U1 as appearing on the schematic
Also check the DC volts on U1 pins 4 and 5, they should be almost the same

[Wallace Gasiewicz]

Also try the variable resistor (POT) Labeled HF1 in the schematic.

[Alex_Baker]

It all measures fine, the voltages on pins 8,9,10, and 14-16 are all within 100mV, same with pins 4 and 5, they measure within 10mV of each other.

HF1 is fine too.

[Wallace Gasiewicz]

Unfortunately it looks like U1 is bad then. Maybe someone else has another suggestion.

Maybe try on this group:        https://groups.io/g/HP-Agilent-Keysight-equipment/messages

Explain your recent findings including your scope pics and see if they have any suggestions

[Alex_Baker]

hmm, I will keep looking.

If U1 really is bad, and I can't find a cheap replacement, then this scope is parts.

[strawberry]

swap diff pairs to see
U1 cant be made discrete

[Alex_Baker]

swap diff pairs to see
U1 cant be made discrete

I don't think I know what you mean.
U1 is a proprietary part, I have no hope of reverse engineering it with the skills and equipment I have, but that does sound like a fun project.

[0culus]

hmm, I will keep looking.

If U1 really is bad, and I can't find a cheap replacement, then this scope is parts.

Sphere Research will probably have it.

[Alex_Baker]

Thanks for the info! At the moment they don't have one, but that looks like a very useful site.

[0culus]

Thanks for the info! At the moment they don't have one, but that looks like a very useful site.

Email them directly to see for sure. Walter has a massive stash.

[Alex_Baker]

I will if it comes to that, thanks!

[Alex_Baker]

Over on the HP-Agilent-keysight group I think we came to the conclusion that U2 on the vertical output board A5 is faulty, not the outcome I was hoping for but sometimes that's how it goes.

That leaves me with two things to sort out:
1) where to find a replacement, making a retrofit is an option thankfully, sphere research has not responded to my email.
2) why did the original fail? any thoughts?

[0culus]

Over on the HP-Agilent-keysight group I think we came to the conclusion that U2 on the vertical output board A5 is faulty, not the outcome I was hoping for but sometimes that's how it goes.

That leaves me with two things to sort out:
1) where to find a replacement, making a retrofit is an option thankfully, sphere research has not responded to my email.
2) why did the original fail? any thoughts?

Walter reads the HPAK list so hopefully he will see it one way or another.

As for why it failed, you'll want to examine the circuit surrounding it. ICs don't usually "just fail*"; it got too much of something. So you'll want to examine all the passives carefully, make sure the voltage rails in the area are OK, etc.

*it can happen, but you definitely want to make sure the circuit isn't messed up and is poised to blow up a replacement!!

[Alex_Baker]

I really cant find anything suspect, all resistors and caps look good. All the caps are little ceramic or mica ones, which are not likely to fail catastrophically. And if something came down the delay line I would expect it to blow U1 not U2.

[Alex_Baker]

No dice on finding a replacement U2 from sphere research.

I think making my own retrofit circuit on a drop in PCB is my best bet, I have the tools and skill to produce a PCB I just need the circuit, which doesn't sound like it is too difficult to find.
I likely won't get to making a replacement immediately but I will start working on it.

I do want to ask how thick the trace should be, the best I can get mine is 1.5 to 2mm, is this normal for a storage CRT? or can some stuff be attributed to the vertical problems.

[Alex_Baker]

This is the only example of a replacement circuit for U2 that I could find on google(somebody may have linked it already)


The person who made it said that it worked quite well but was uncertain about the bandwidth. I would like to get as much of the 275MHz bandwidth that I can, but I am perfectly willing to make a compromise at 150 or 200MHz, which is still faster than I will need for quite some time.

They said that they used mosfets with the part number D16NE, which doesn't seem to exist. So my question is what specs will be important when picking out a mosfet to use for this project?

[Wallace Gasiewicz]

https://www.datasheetq.com/search.jsp?sWord=D16NE&op=i

N - CHANNEL 60V - 0.07Ω - 16A DPAK/IPAK STripFET™ POWER MOSFET