Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
Fix or Part Out a Tek 475A
|
I've got a 475A that I acquired for parts (for a 475) but it looked like an interesting repair project, so I've been trying to fix it instead.
I managed to fix the low +110V rail simply by unsoldering and re-soldering capacitor C1489. With that surprisingly easy repair under my belt I ventured to the next obvious malfunction: the beam intensity adjustment. I checked the pot and, while it is stiff, it seems to be working properly (the voltage on the center tap swings fully from -8V to +15V when you rotate the knob), so I decided to try checking the components further along toward the CRT. After careful examination of the schematic and the circuit board images in the service manual, however, it appears that most of the components are underneath the high voltage shield, and I'm a little bit scared of that section. I almost feel comfortable removing components from that area with the power off (almost, but I know that the CRT can retain a charge for quite a long time, so anything directly connected to the CRT could still be "live" even with the power off), but I'm not willing to poke around in there with power on. The circuit in question is only subject to voltages in the range of 100V, but it's all deep in the HV section, so I know that there are KV potentials just waiting to ambush the unwary/clumsy adventurer, and I don't want to be that guy. It looks like someone else may have tried to repair this same fault, because one of the screws for the HV shield has been stripped, which could indicate that the fault is actually quite resistant to repair. I was kind of interested in having a working 475A, but I understand that it's not really that much better than the 475 (and in some ways it's worse), so I'm thinking that this is the point at which I should call this a confirmed loss and just use it for parts. Am I being wise and prudent in my caution/fear of the HV section, or is this still something that I should consider trying to repair? |
|
Do not fear an unpowered scope¡¯s hv section. The stored energy is too small to be lethal to someone in otherwise good health. A zap would be unpleasant but that¡¯s about it.
toggle quoted message
Show quoted text
If you want to avoid even that unpleasant possibility just discharge the hv. Again, I am talking about an unpowered (unplugged) scope. Don¡¯t let a fear of a small hv bite inhibit you from fixing the scope. I¡¯ve been zapped countless times by much higher voltages, and aside from uncontrolled drooling and a limp, I¡¯ve suffered no ill effects. Cheers Tom Sent from my iThing, so please forgive brevity and typos On Nov 21, 2020, at 13:29, Jeff Dutky <jeff.dutky@...> wrote: |
|
You sound like a crusty old elevator constructor technician! :)
toggle quoted message
Show quoted text
Bruce Gentry, KA2IVY On 11/21/20 4:56 PM, Tom Lee wrote:
|
|
That¡¯s pretty much my job description!
toggle quoted message
Show quoted text
:) Tom Sent from my iThing, so please forgive brevity and typos On Nov 21, 2020, at 14:38, greenboxmaven via groups.io <ka2ivy@...> wrote: |
|
Thanks for the encouragement, I was feeling bad about having made some progress only to give up now.
I'm only a crusty old software engineer, but I noodled around with electronics in my youth, and got zapped a few times (never with anything in the KV range, though), but I would like to avoid that experience if at all possible: how do I discharge the HV section? Is it like a TV CRT, with the shrouded connector on the side of the tube (I think I'd need to remove the even scarier cover on the CRT section to do that) or is there another point I can discharge from? Also, what should I use to do the discharge (my grandfather would have just used a fat screwdriver to discharge a CRT, but I have doubts about the wisdom of that technique)? (I'm checking the 475A service manual for advice on discharging the HV section, but there's no index). |
|
It is the same procedure as for a crt TV, just as Dad did back in the day. I frequently see online advice to use a series resistor for the discharge but often with a nonsensical quarter- or half-watt device. The flashover voltage of these little resistors is well below a kV, so they¡¯re doing nothing in that application. A resistor isn¡¯t truly necessary, so your father had it right. Back in my TV service shop days, we just used a screwdriver gently slid under the anode cup, and held there well after hearing the snap, to ensure a good solid discharge.
toggle quoted message
Show quoted text
Tom Sent from my iThing, so please forgive brevity and typos On Nov 21, 2020, at 15:25, Jeff Dutky <jeff.dutky@...> wrote: |
|
A grounded screwdriver, just in case that wasn¡¯t obvious.
toggle quoted message
Show quoted text
Sent from my iThing, so please forgive brevity and typos On Nov 21, 2020, at 15:35, Tom Lee <tomlee@...> wrote: |
|
The normal way to discharge Tek CRTs with Alden connectors, is to carefully pull the CRT end of the connector out, and discharge the tip to the chassis. Problem is, this isn't always easy, because the connector halves are often stuck together pretty well - especially if one part is silicone rubber. Also, it's usually in tight quarters, with fairly short leads. First twist the halves to break the bond, then hold the outer part firmly and twist and pull the inner part just until it unsnaps, and withdraw it slowly. The real trick is to not be nervous or shaky, and accidentally let go of the CRT lead - hold steady until you can get the exposed pin to contact only an open area of chassis or shielding. If you slip and let it loose on the way out, the lead may flail around and zap your fingers, or worse, land in some nearby circuitry. Dexterity and control are crucial - like opening a bag of chips without having it suddenly shred and fling the contents everywhere. Practice makes perfect - you can eventually do it with one hand, without even looking.
Ed |
|
So I've gone ahead with trying to fix the 475A further. I examined both the block diagram of the CRT control system, and the schematic for the beam intensity and blanking amplifier. There are 7 transistors involved in the beam intensity amplifier, and I located each of them on the circuit board.
With the unit powered off and unplugged, I carefully extracted each transistor, one at at time, and ran them through my component identifier tool. I found two parts that the tool did not see as present in the test socket (I assume that these are blown open, Q1352 and Q1358), one that triggered the self-calibration function (assumed to be blown short, Q1344), and one that reads as a PNP when the parts list says in should be an NPN (Q1338, I don't know what to make of that: I've seen that the board images are occasionally wrong, but if the schematic and parts list are wrong, what can I do?). I replaced the shorted transistor straight away (it's a bog standard 2N3906), but the others are more difficult: Q1338 is an NPN SKA6516 according to the service manual, but the device is marked identifies as a PNP transistor, the part is printed with a Tek part# "151367 TI 877" or "TI 811" it's hard to tell from the style of the printing. I'm tempted to leave this one in place until it is proven to be a problem. Q1352 is an NPN 2N3501 according to the service manual, but the manual also says it is "selected from 2N3501". Without knowing the selection criteria, how can I possibly replace it with a comparable part? Q1358 is a PNP MPS3640 according to the service manual. I have looked at Newark, Arrow, Mouser, DIgiKey and JameCo for replacements, with very little success. I have found an SMD part that appears to be the same number (MMBT3640) but it's power dissipation is only a fraction of the MPS3604 in a TO92 package. I'm guessing (from an abundance of ignorance) that the headline spec is Ft of 500MHz, but I know that I would need to exceed the other specs as well (not so hard for Vcb, Vce, and Veb, but power dissipation seems to be hard to match, and I haven't really looked at collector capacitance). I know that some of the "selected" parts are selected in groups of two or more, usually to have matching characteristics, or characteristics in some precise ratio. I don't see any indication of what that criteria would be in this case. So, my questions are: 1. What should I do about Q1338 which seems to be a different part than what the manual specifies? 2. How do I "select" a new 2N3501 to replace the blown part? 3. How do I choose a replacement for the MPS3640 that doesn't seem to be manufactured anymore? Thanks -- Jeff Dutky P.S. I have managed, thus far, to avoid getting zapped. |
|
You don¡¯t say what the symptoms are. No trace? Not found by a press on the beam finder button? You can work on the HT section by taking off the metal screening plate (5-6 screws). If you look around the front left side, you will see test points for the cathode (-2450 V) and first grid (-2530 V). They are close to thick wires that route to the CRT socket. Measure these voltages with a meter (full scale 3000 V) using well insulated probes. You can also measure the grid-cathode voltage with a plastic multimeter; should be around -80 V and vary with intensity and and grid bias pot settings.
On switching off the scope these voltages decay fairly quickly, but it is best to be sure by briefly grounding the cathode test point. The accepted technique when probing high voltages is to always have one hand behind your back, a zap through the whole body to ground is not so bad as a zap from one arm to another. If you have good HT, it is worth switching on the scope in a darkened room to see if you have a green glow from the front of the CRT, which would indicate that the tube is OK, but that the beam is too displaced to form a spot or trace (wrong signals on the X Y plates, in which case it is best to troubleshoot the drives to these plates. The X plates are driven by the red and green wires going through a hole in the main board in front of the HT section (position 7E on A9 board). The Y plates can be probed at the resistive T network (R497) connected to the front two Y plates. Be careful probing this part which is a little fragile and very hard to replace. The front two Y plates are connected to the back two Y plates inside the CRT. The back two Y plates are driven by the Y output custom IC amplifier U470. Mine had gone bad and the spot was deflected way off center. If you suspect the THT circuit supplying the post deflection anode (around 15 kV), you can test it by pulling out the white plastic plug holding the fat red wire. This voltage is dangerous and takes a while to decay after switch off. TV repairmen know how to ground this by sliding a grounded flat blade under the rubber boot on the front of the CRT. You can test the THT by poking one end of an automotive ignition wire connected to a spark plug into the THT socket. If you get a good spark chances are the THT is good enough to produce a bright spot. If the HT and THT are OK, I would check the waveform at TPs 1364 and 1366 to make sure you are getting enough voltage to unblank the beam in sync with the timebase. Reseating the transistors in this circuit may be needed. Simon |
|
There are a number of symptoms, which are a bit vague:
first, and most obviously, there is no response to turning the beam intensity knob; the trace stays the same brightness no matter what the knob (or pot) position. I verified that the voltage on the center tap of the beam intensity pot swings between -8V and +15V when you rotate the knob. second, there appears to be a "ghost image" of the trace at the ground level. At first I had thought that the trace was "wrapping around" the screen, but now I suspect that I am seeing the retrace that is not being properly blanked. third, I noticed that the A INTENS mode didn't show a clearly intensified region, but seemed to respond to the delayed B trigger correctly. Again, this points to a problem in the beam intensity amplifier. Other than these malfunctions I have full, clear traces on the screen, and they respond to the focus adjustment. This leads me to suspect that the high voltage section is working mostly properly, and that the tube is in good condition (I've had absurdly good luck buying "for parts" scopes from eBay. Everything has powered up and shown traces on the screen. The only malfunctions have been this 475A, which is still being fully diagnosed, and a 2215A that had a blown capacitor in the horizontal position circuit. I was expecting much less functionality, and much less guilt about disassembling the non-working scopes for parts). Also, I don't currently have a "well insulated probe" for my multi-meter (it's on the list of things to get), so I'm not willing to poke at the 1250V test points. Fortunately I don't think that the HV section is suspect, outside of the beam intensity amplifier. As I said above (maybe while you were posting your message) I already went in and pulled each of the transistors in the beam intensity amplifier, one by one, and tested them in a component tester. Three of the transistors appeared to be bad, one acted like a short circuit, and the other two acted like fully open circuits. When I posted my reply above, however, I had been looking at the wrong service manual (I had mistakenly opened the manual for the 475 rather than the 475A). One of the two blown transistors (Q1352) does not have a non-Tek part number specified; it is listed only as 151-0407-00 from manufacturer 80009 (Tektronix). The same part in the 475 is listed as a 2N3501, but I have no idea what help that might be. I'm something of a novice when it comes to transistors, so I don't really know what characteristics I should be looking at, or even what most of the characteristics mean (I can make a guess, but it's not what you could call an "educated guess" other than that I know the difference between voltage, current, power, and frequency, and I have heard of most of the terms involved with transistors). |
|
You can check the operation of the Z axis amplifier by looking at the waveform at TP1366 or TP1364. I would have most faith in the circuit diagram and what is marked on the transistor in the circuit unless you think someone else has been there. These are high speed switching transistors and you should be able to find equivalents from an internet search. Pinouts can vary between manufacturers. The power supply is 110 V so some may require higher Vce.
Simon |
|
On 2020-11-22 2:32 a.m., Jeff Dutky wrote:
... One of the two blown transistors (Q1352) does not have a non-Tek part number specified; it is listed only as 151-0407-00 from manufacturer 80009 (Tektronix). The same part in the 475 is listed as a 2N3501, but I have no idea what help that might be.For a start, you will need the Semiconductor Cross-Reference, which you can find here: --Toby |
|
Jeff,
I have repaired many of this series of scopes. I would say that once you get the proper transistors in place, that you should either fix the issues or narrow the problem down to a dirty switch, bad Pot or a bad component in the DC Restorer circuit. Here is a valuable resource that can assist in identifying specific characteristics of "selected" components, as well as standard ones. This is the TEK "SEMI-Conductors: Common Design Parts Catalog". With some searching you can find the specifics of many of these "selected" components. This also supplies alternate parts information, through a pretty substantial TEK Part to Mfg. Part interchange as well. This covers most of the TEK semi-conductors of the era, not just transistors. . I am not sure if you have discovered TEK Wiki or not. Well worth your time to explore. Main Page Here: Good Luck! -- Michael Lynch Dardanelle, AR |
|
Michael and Toby,
Thanks for the pointers to TekWiki. I was aware of it, but there is a lot of depth there, and it helps to have guidance (e.g. I had found the Semiconductors Common Design Parts Catalog through Google, while searching for this specific part, but I had not understood that it mapped Tek part numbers to commodity parts). This doesn't address my mistrust of the service manuals, but it does fill in many of the blank spots. Thanks again -- Jeff Dutky |
|
On Sun, Nov 22, 2020 at 06:35 AM, <toby@...> wrote:
I think, Q1352 is a high voltage, silicon, NPN, video driver in a TO-39 can... which makes sense to me because it is used in the Z-axis amplifier. Tektronix says you can use the similar? 2N3923, at least according to the High Voltage NPN Table in Section 6, page 6-6, of the Tek parts cross reference. Supposedly, the NTE154 is a substitution for the 2N3923, and might be more available. But NTE154 is a sub, of the sub... so caveat emptor, on that route. These aren't really expensive transistors... less than 10 USD? ... so maybe get a couple. The 151-0407-00 transistor is used in some other Tek too... so if you poke around on the Internet people were selling that Tek part and the 2N3923 too. |
|
I've ordered replacement parts for Q1352 (151-0407-00 / 2N3923) and Q1358 (MPS3640). While I'm waiting for those to arrive I figured that I would evaluate the passive components of the circuit as well, on the basis that if the transistors failed, then maybe some of the stuff around them also failed. My plan is to pull the transistors and check with a multi-meter between the individual pin sockets for blown/shorted components. Based on the schematic I can get unambiguous measurements on every path.
I can do simple voltage/resistance checks, but it looks like the important properties of these diodes in this circuit are going to be dynamic (e.g. how fast they cut in/out) and I don't really know how to do those measurements (in or out of circuit). I figure it's worth my time to find out if anything is grossly failed (open/shorted), but I'm not sure about testing beyond that level. As I have said, I am a novice when it comes to semiconductor circuits: I have a basic understanding of RLC circuits (what I learned in undergrad physics), but my mastery of diodes and transistors is, at best, rudimentary (again, what I learned in undergrad physics). Am I correct in assuming that so long as I replace any shorted components there is little risk of the re-powered circuit immediately frying the new transistors? |
|
While I'm waiting for the replacement transistors to arrive, I have proceeded to try to check the other components in the beam intensity amplifier circuit. After a careful review of the schematic it looks like I should be able to get clean readings for most of the components after removing the transistors from their sockets. I pulled the five transistors and started testing the nodes between them. I think I've found a couple of things that look shorted that should not be shorted. At least one diode is shorted out, and it looks like someone else might have had a try at reworking this board. I'm not sure but it looks like they might have solder bridged some pads that shouldn't be bridged.
It also looks like a some resistors are out of spec (R1354 and R1356), though not obviously damaged in any way: there are two parallel resistors across the collector and emitter of Q1354 that don't have anything connected between them (they look like a voltage divider, but nothing is connected at the midpoint where the divided voltage would be pulled off). On the actual circuit board these two resistors are stood on end next to each other with their upward pointing leads soldered together. I have no idea what this is supposed to achieve, but the resistors measured values (9.5K each) do not match the values specified in the schematic (7.5K each), nor do they match the values indicated on the body of the resistors (6.5K, which also do not match the schematic). If I'm going to need to remove and replace multiple components, as well as fix previous rework, I think I need to take the main interface board out. It looks like I basically have to disassemble the entire unit in order to remove the main interface board, which is more work that I was bargaining for. I'm willing to do that amount of work, but the risk of causing further damage is pretty high, no matter how careful I am about the endeavor, and there's no guarantee that I will get a working scope out of the process. As I said, I'm not averse to putting out the effort, and I would certainly learn things in the process (that's basically the whole reason I'm doing this, as a hobby to keep me occupied during quarantine), but this does seem to be spiraling out of control, at least a little bit. -- Jeff Dutky |
Michael W. Lynch
to navigate to use esc to dismiss