HP 1727A oscilloscope with multiple issues.

[0culus]

https://www.sphere.bc.ca/download/hp_xref-free.pdf

[Alex_Baker]

That's the cross reference sheet I used. I found no references to a transistor when I searched PH3563-5, google only came up with other cross reference sheets!  my gut wants to tell me that either 2N2369  as suggested by dzseki or PN918 will do the job.

I don't think the original transistor is a metal can one, because according to the parts list there are several other of the same transistor on A13, and they are plastic. Although since my scope seems to be a prototype all bets are off. I am sure a metal can will work though. 

[artag]

I did a little research and it seems that there are not too many of these things out there, I did find two other examples of scoped with the same second smaller trace, or at least it appears to have it. The first link is an ebay listing for a 1742A not a 1727A, however it looks like it has a smaller trace in the middle(the bright part), however it doesn't look like the 1742A is a storage scope, so perhaps the second trace is not caused by the storage bits?

I have a 1725A, a non-storage scope that's also 275MHz. On that one, there is sometimes a small bright region on the trace. It's due to the delayed timebase : unlike a Tek 475 where the dot suddenly accelerates to the second timebase speed, the HP way is to generate a full-width trace that acts like another channel (you can shift it vertically) and brightens the main trace to show the zoomed section. I guess they do this by alternating the trace.

If you want to disable it, turn the delayed timebase to the same speed as the main timebase. On my scope, once you've done this the two knobs are locked together and have to be intentionally unlocked to get a delayed trace. You should also find the delay setting (the vernier control) will change the position of the bright-up within the main trace.

[Alex_Baker]

Thanks Artag, that's good to know 

[artag]

It is also huge...I'd say it's stretching the definition of "portable oscilloscope". 

Of course it's portable. It's got a handle, hasn't it ?
Non-portable scopes are a two-man lift onto their cart

I swear engineers are getting weaker !

[cosenmarco]

I found no references to a transistor when I searched PH3563-5, google only came up with other cross reference sheets!

What about https://www.mouser.com/datasheet/2/149/PN3563-889953.pdf ?
This transistor is used as pre-driver in the push-pull output amplifier. As long as max current, ft, power and hfe range are comparable or better you can replace it.
Being that in a sort of push-pull amplifier, you may also want to replace with the complementary of Q13 (Q13 is according to manual a 2n4917 - https://alltransistors.com/transistor.php?transistor=4299) I can't find right now what transistor is supposed to be the complementary.

[Alex_Baker]

Does Q14 have to have a ft of 600MHz when Q13 only is rated at 200MHz in the manual? could you get away with a 200MHz replacement for Q14?

[Alex_Baker]

Ok now I'm confused 

The schematic for A13 shows Q13 as PNP and Q14 as NPN right? I just noticed that the parts list for A13 lists both transistors as NPN. I am going to go off of the schematic and assume that they meant PNP for Q13 though.

I found 2N5400 as a possible replacement Q13, which is PNP. https://www.mouser.com/ProductDetail/610-2N5400-PB, Do you think it will work?

[cosenmarco]

The schematic for A13 shows Q13 as PNP and Q14 as NPN right? I just noticed that the parts list for A13 lists both transistors as NPN

I do see in the parts list
A13Q13 1853 0089 Transistor PNP SI PD 200mW 07263 (mfr. code) 2N4917 (mfr part n.)
A13Q14 1854 0092 Transistor NPN SI PD 200mW FT 600MHZ 28480 (mfr. code) 1854 0092 (mfr part n.)

Although it's very badly scanned and not so readable in my case. I'm using official service manual from keysight site.

Q13 only is rated at 200MHz in the manual? could you get away with a 200MHz replacement for Q14?

are you maybe confusing with PD ? 200Mhz would be too low I suppose.

Code: [Select]

2N4917 Datasheet, Equivalent, Cross Reference Search
Type Designator: 2N4917
Material of Transistor: Si
Polarity: PNP
Maximum Collector Power Dissipation (Pc): 0.2 W
Maximum Collector-Base Voltage |Vcb|: 30 V
Maximum Collector-Emitter Voltage |Vce|: 30 V
Maximum Emitter-Base Voltage |Veb|: 5 V
Maximum Collector Current |Ic max|: 0.1 A
Max. Operating Junction Temperature (Tj): 125 °C
Transition Frequency (ft): 450 MHz
Collector Capacitance (Cc): 5 pF
Forward Current Transfer Ratio (hFE), MIN: 150

I think the 2N5400 is going to be fine. At least it won't blow up.

[Alex_Baker]

PD for Q13 is listed as 1W in the manual, and FT at 200MHz.
PD for Q13 is listed as 200mW in the manual, and FT at 600MHz

I attached a picture of my manual for that section.

[Alex_Baker]

Mouser doesn't have anything that matches 2N4917, and filtering by the specs you gave 2N5400 is the only one I can buy small number of, unless I want 5000 little transistors laying around. 

I think 2N5400 is the best choice for me at this point.

[Alex_Baker]

one other question I have is if I can replace U1 on A13, which is a CA3140T op-amp, with a LF356H op-amp. I still need to test U1 but I have a hunch that I popped it when I switched its +15V to the +53V like it should be in the schematic. It is most likely dead because it is supposed to have a 30V zener (VR3) to reduce the voltage across it to just 23V, but that Zener was replaced with a wire link in my unit, so the op-amp got the full 53V across it. The reason I want to replace U1 with LF356H specifically is because I already have then. 

On that subject, do you think I should undo all the little hacks/mods on A13? or should I leave it the way it was when I got it. The problem is that I have no idea if the original setup with all the little mods even works.

CA3140T: https://pdf1.alldatasheet.com/datasheet-pdf/view/66338/INTERSIL/CA3140T.html
LF356H: https://pdf1.alldatasheet.com/datasheet-pdf/view/557688/TI/LF356H.html

[cosenmarco]

if I can replace U1 on A13, which is a CA3140T op-amp, with a LF356H op-amp

I think you can try to replace it with the LF356H

do you think I should undo all the little hacks/mods on A13?

Not really sure... I think you should compare your manual with the one from keysight which hopefully is the last version, also have a look at included manual modifications and errata which may be included in that more recent manual. Also I would keep it as last resort and maybe do that in a controlled way like have the equivalent of that portion made on a protoboard and try replacing only that portion of the PCB while keeping the original components, comparing signals, measuring in a systematic way, trying to compare the measurement with a model I have in my mind of what the circuit should do or better with a simulated circuit.

[Alex_Baker]

I ended up testing the op-amp on a breadboard and surprisingly it seems like it survived having 53V put across it! impressive.

As for the mods I got parts to undo them If I want to in the future but it begs the question why it was modified in the first place. I am tempted to reverse engineer that part of the circuit is EasyEDA and compare it to my manual and the online manual.

[Alex_Baker]

Ok, here is what I drew out in EasyEDA, on the left is what I have and on the right is what the manual has. (refer to the manual for resistor/diode/cap numbers)


I don't know enough to tell if this circuit would work as it is, but I do know that with a supply voltage of 15V VR4 would not be doing anything, which seems like it would be a problem. VR3 is replaced with a wire link on my board, but I don't think that would keep the op-amp from working since there is still 15V across it, and the base of Q7 is at 0V referenced to the op-amp anyways.

[andy2000]

I recently repaired and sold one of these.  It's an unusual oscilloscope to use.  Unlike the Tektronix storage scopes, this one doesn't have a non-storage mode.  The setting of the brightness control is critical.  Even a little too bright, and you can see a shrunken phantom image.  A little more too bright, and the whole screen becomes a mess.  I didn't find it to be a very easy to use scope.  Definitely read the manual if you haven't already because if you treat it like a normal scope, it won't work well. 

Mine had a leaking cap in the HV supply that caused HV regulation problems. 

[Alex_Baker]

HI andy2000, what would bad HV regulation look like? just curious in case I have a similar problem. I don't have the proper multimeter probe to measure 2700V though.

intensity settings causing the shrunken trace makes sense, as the intensity control doesn't seem to work at the moment. The brightness control does something, although not very much. Once I get the parts in from mouser for A13 I think it is time for a project recap so everybody on the forum an myself knows where I am at as far as what issues have been fixed and what issues still remain.

I honestly do not know what half of the controls on the front of this scope do.

[andy2000]

It caused brightness issues that got worse as it warmed up until it was completely unusable.  Check the electroytic caps on the HV board since I wouldn't be surprised if it was a common problem.  Even when it was working properly, the settings of the brightness and intensity controls was extremely critical. 

[Alex_Baker]

I have heard that brightness and intensity is important.

The HV board only has a couple of little axial electrolytics, the rest are polyester or something similar it looks like, I doubt the polyester ones are leaky but I will still test them.

[cosenmarco]

So, I see a couple of things with this circuit that make me not want to have to do with it. I don't even know if I want to comment because I may give you wrong advice so take all of this with a grain of salt.

I don't know enough to tell if this circuit would work as it is, but I do know that with a supply voltage of 15V VR4 would not be doing anything

Here I believe that the diode is there for protection. The "Accel" signal comes directly from CRT so it may float due to strain charge to levels which would fry the opamp. ... at leat I think... the opamp input should never go lower than V- minus 0.5 but there it seems it may go as low as -3 V... so  perhaps they did just see no harm in leaving the Zener there and didn't want to take the time / effort needed to remove it.

I think this circuit works in a broader context. The feedback loop of the opamp involves some elements of the CRT itself and additionally it has some components (R18, C3) which are factory picked, so they must be selected depending on the CRT and / or other components.

It seems to me the additional zener you don't have was used to float the circuit to a certain level so that the input signal coming from CRT "Accel"  could be referenced to the level set by the "Current limit" pot and at the same time integrated.
The output  of this integrator sums up in current at the junction between CR3 and CR4 with another current which depends on Time/Div setting making up the amount of current which is sunk by Q12 which ultimately modulates the output gate amplifier.

The goal of this would be to put a limit on the writing gun current. I think they would choose at factory components by measuring the voltage at teh "Accel" grid given a certain cathode current for the specific CRT and in this case they perhaps saw that the mod was necessary to keep the circuit working within tolerance.

As a bonus you get a signal which compensates the focus at various intensity levels (thus "auto focus").

I think messing up with this circuit poses the risk that this very same circuit tries to mitigate: burning the storage mesh with excessive current: If I understand correctly the storage mesh is made of non-conductive material so it could be burned by a beam which is too strong. As of which value the current should be limited to... I have no clue and must be some parameter on the datasheet of the CRT which of course is not easy to obtain I guess. Even though you have the data, measuring the relation beween Accel grid voltage and cathode current is tricky due to the high voltages involved (but not impossible).

[Alex_Baker]

I got my parts in today to replace the transistor on A13, along with some new zeners and rectifiers for the LVPS board just as preventative maintenance.

I still need to do some testing to see if replacing the transistor fixed anything but I figured I would give you an update since I have been quiet for the last couple of days, I was just waiting on parts.

[Alex_Baker]

Update, there is a recurring thing that has happened multiple times today but once I get out my camera I cant reproduce it  . what it will do is it will start up normally, and then make a quiet "pop" noise toward the front of the CRT, at which point it will lose horizontal drive and the greed background glow.

I don't think replacing Q14 on A13 did any thing, there is still no signal on TP1. Another thing is that now when I turn it on the main trace is not visible at all, only the second one(first picture). I think what is happening is what you can see in the pictures I posted on the first page of this topic, so I think the trace is still there it's just not visible. Too add onto that though I observed that if I unplug the connector J3 on the LVPS (upper right corner), which is the coliminators and lenses, the green background goes away and you can see the main trace(see second picture). I did just replace the rectifiers and R54 on A16, so I need to make sure the coliminators and lenses are adjusted.

I know its a lot to take in!  but any ideas about some of these symptoms would be helpful. 

[Alex_Baker]

I got a picture!

I dont know exactly where the "snap" comes from since it is kind of random.

Another thing I should mention is that the collector mesh on my scope runs at 150V, instead of 120V like you see in the online manual.

[cosenmarco]

Hi Alex

I don't think replacing Q14 on A13 did any thing, there is still no signal on TP1

but you got a signal on the base of Q13,14 right ? So it's just the amplifier stage not working correctly? I still think that you have to get the gate signal up and running before troubleshooting further but of course now...

hat it will do is it will start up normally, and then make a quiet "pop" noise toward the front of the CRT, at which point it will lose horizontal drive and the greed background glow.

I think you may have an issue somewhere related to some capacitor becoming closer and closer to a short and one of the U1-U3 in the A16 LVPS detecting over current condition and cutting some voltage off. It is interesting to see, that even then you have 2 dots on the screen instead than just one. You still apply the input signal? If yes, then it might be the vertical still working so if there is one of the low voltages in the horizontal section which is not in the vertical, taht might be your voltage.... or just measure all of them when it happen. Also the dots are at maximum brightness.. this is compatible with my theory that gate signal is very relevant and needs to be fixed.
Also... don't leave the scope running for long time with such bright traces... the tube may get damaged.

[Alex_Baker]

I don't think there is a problem is the LVPS.

I figured out that the "snap" noise was coming from the two series neon bulbs on the output of the storage mesh. I also learned that when the snap happens and the screen goes dark, it just put itself into "store" mode, so the store LED lights up. I can then get the screen back by pressing the display button. This almost sounds as if something is arcing within the CRT, but I hope not! 

The gate output looks as it should according to the manual, it just lacks the 70V peak amplitude that the adjustments section specifies, so yeah probably the amplifier stage. I tested the output transistors in circuit with a meter but I guess I will just have to take them out and test them properly. One of the output transistors gets pretty hot while the other does not.