I've got the service manual to the Zenith DT2 display (includes a linear PSU); dated 1979 and identified as part of the D12 series of monitors. The HOT is a Zenith 121-1039, which, according to NTE is replaceable by NTE 379. http://www.vetco.net/catalog/product_info.php?manufacturers_id=14&produ… Dunno if that helps or not. --Chuck

On 11/03/2014 07:38 PM, drlegendre . wrote: No problem--and you didn't even have to argue with Angus to find out. :) If you'd like, I can scan the schematic and parts list, if it would be helpful. --Chuck

There are basically 2 common reasons for a HOT to run hot (!). One is overcurrent. This generally means something is loading the flyback too heavily. Remember that this is a high voltage circuit and components may test fine on the DMM at a few volts and break down when in the actual circuit at a few hundred volts. The other common reason is that the device is being turned on or off too slowly. The horizontal output stage is a SMPSU (with extras). The transistor should be hard on (saturated) or cut off. If it takes too long to switch between those states it will disipate too much power and run hot. I should have been clearer. A power transistor, such as the HOT, often has a low beta. If the circuit requires a component with a high beta then often a darlington pair is used. It would be useful to know what the original HOT was. Have you tried testing the B-E junction of the old part with the diode test function of you DMM? what forward voltage do you get ? If it's 2 diode drops (about 1.4V), it's very likely to be a darlington device. Remember that voltage gain isn't the whole story. You might have similar voltages on input and output but a larger current in the latter circuit. That would be an ampifier, giving power (and current) gain. Basically you resonate one of the flyback windings (out of circuit) with a capacitor. Apply a pulse signal to it (often a little oscillator and a HOT-type of transistor is used). And look at the waveform across said widing with a 'scope. You should see a decaying sine wave. If it decays very quickly (maybe not even one cycle) then something is taking energy from the resonant circuit, most likely a shorted turn in the flyback You will never find a shorted turn by DC resistance measurements. The change in resistance is less than the manufacturing tolerances (due to the wire diameter not being exact), contact resistance, etc. -tony

Looks like we're mostly over the hump with the list breakage, so it's time to post an update on this. Our friend Chuck cracked the first nut, by essentially solving the mystery of the HOT. Turns out that the "1070" marking is an abbreviated Zenith in-house part number; the full version is 121-1070. Also turns out that the 121-1070, like the Zenith 121-1039 in his 12" display apparently crosses to an NTE 379, which itself may effectively be a BU408. Again, a big round of thanks for Chuck on this work.. score one for the flame of knowledge. So based on the specs of the NTE 379 with an avg, beta of 18-24 (min. 6/8 / max. 30/40), the extant HOT (1070) with a measured beta of 11-12 may actually be in-spec, albeit on the very low side. But an average of ~18-24 is a far cry from the 150 of the unit in-use.. so that leads me to a question, of course. The HOT basically acts as a switch in this application - it's really an on/off proposition vs. linear amplification. So why is it that the original HOT wasn't able to transition to a sufficiently low C-E resistance to do the job? For some reason, I have the odd feeling that this is due to a lack of base drive, rather than an excessive load imposed by either the HV supply, yoke and/or the derived tertiary supplies (15V, 50V). Here's my line of logic: The beta / Hfe / gain doesn't really matter - if the HOT saturates, it saturates and falls to a very low C-E resistance. The sub part (NTE 375, which is a +vertical+ opt. part) is obviously saturating, as we're getting plenty of deflection & output from all the supplies. If the original 1070 doesn't produce the same results, then it seems the the reasons are pretty limited... It can't be due to excessive loading by the tertiary or HV supplies, as those can only rise as high as a saturated HOT will allow. +It's not as if the NTE 375 has any significantly lower "on" resistance than the 1070 could, if both are saturated, right?+ So then, if the various supplies & deflection are OK with the 375, there's nothing a "better" or "stronger" HOT could do to increase those levels - see my logic? This points to only one (well, two) possibilities: Either the base drive is inadequate to saturate the low-gain 1070 (yet fine to saturate the NTE 375, with 10X the gain) -OR- there is some defect in the 1070 itself that's squelching what would otherwise be an adequate base drive - i.e. its input impedance has somehow fallen very low, and way out of spec such that it's killing the drive. Your comments are not only appreciated, but basically required.. ;-) On Tue, Nov 4, 2014 at 12:39 AM, Tothwolf <tothwolf at concentric.net> wrote:

Tony, All, On Wed, Nov 5, 2014 at 11:47 PM, tony duell <ard at p850ug1.demon.co.uk> wrote: There was more to it than that, but I described it many posts ago, and you probably missed the details.. here they are again, for anyone who might be curious. It's not that there was a solid, bright vertical line - the image consisted of the contents of the display (an ACSII block message box, plus a line or two of text) squashed into the vertical. So there was a short, dashed vertical line about 1.5" high, in the upper third of the screen. It was dim, and became more dim over the course of a minute or two - eventually disappearing totally, appearing to slowly slide upward past the upper limit of the screen. More on "disappearing" coming up.. I now realize that it's important. re-install it as part of my next-step troubleshooting. I then disconnected the tertiary supplies (15V & 50V) by lifting leads of W3 & CR104, respectively. Guess what? The 50V supply came up to 44V (the highest I've seen it).. then the stupid stuff started.. I re-connected the 15V supply, and checked the 50V - still good. Then re-connected the 50V supply - and it's still good.. AND the d*mn display is working!! Now how obnoxious is that? Oh, and the HOT was running quite cool, barely over 100F. But not for long.. soon, the display faded out and the 50V supply crashed to 20V. Ack, it's an intermittent..! Replaced C114 & C401 (the stabilizng caps for the 50V & 15V supplies) just for drill, to no effect. So this leaves a couple of possibilities.. There's an intermittently shorting device in one of the tertiary supplies -or- the HOT itself is intermittent. Also possible that there's an issue with the base drive dropping out, but I'm putting my money on the HOT itself. The base drive question can be eliminated with a scope, but I'm going to replace the HOT in any event. Anybody got a spare laying around?

Hi Simon, All - for an update (below) On Thu, Nov 6, 2014 at 1:16 PM, Simon Claessen <simski at dds.nl> wrote: You know the saying.. when all you have is a hammer, everything looks like a nail.. ;-) IOW, it's largely due to my lack of experience.. I'm sticking to the stuff I know, and when it comes to analog video, the horizontal output section is about the only area I have any understanding. Also, I do know a horizontal issue when I see one (vertical line) and when you add-in the low tertiary supplies, the HOT is at the core of it. As for the rest of the caps around the board, I've checked them all with my eyes (for liquid leakage / blow-ups), an ohmmeter (for static shorts / electrical leakage) +and+ a quality in-circuit ESR meter (Capacitor Wizard) and can't find a bogie. Not to get twitchy, now, but I heartily challenge your assertion that transistors "either work or don't".. transistors have more than the two (hard short / hard open) common failure modes - it's in the literature. ** General update commences here ** But the thing is, I've eliminated the tertiary supplies.. by lifting the diode / wires as I did, they are disconnected from the horiz. section. And in this state, the 50V supply still crashed on me, did so this afternoon - and yes, I have replaced all caps connected to the 50V line. It is clearly a thermal intermittent issue - things are OK when cold, and go haywire after a few minutes time. One reason that I'm still suspecting the HOT, is that the entire time I had the NTE 375 replacement installed, the 50V supply never crashed. No, I didn't run it for that long, but based on how fast things go bad (when it warms up) I really should have seen something when the 375 was in place. Oh, and to eliminate questions about shorts in the horiz. yoke, I pulled the connection while the 50V supply was in the crashed state - and the supply rose by perhaps a volt or two - so the yoke is clearly not the source of a short-circuit. I still suspect the HOT or components to the left (aka base drive). I need to let the system get good & cool before I hook up the scope to watch what happens to the base drive when the 50V supply crashes. if it's stable, then it's either the HOT or possibly the flyback diode or the (what's it called?) safety cap? The "special" 400V cap in parallel with the flyback diode? I'm running out of stuff it could be.. of course, there's always the horiz drive transformer, ugh.

** Update - I think it's nailed ** I let the unit cool down sufficiently such that things were "working" again - but this time, I was ready. I had the base & collector of the HOT instrumented, as well as the first cap in the 50V supply. Recall that both 15V & 50V tertiary supplies are otherwise disconnected from the horiz. section where they are produced. With the 50V supply running fine at 44V, the signals at the HOT base measured 212mV P-P, and the collector 15V P-P. This ignores any HF spikes, I'm just concerned with the significant power region of the signals. After a few minutes, the 50V supply crashed.. and guess what? The HOT base signal actually increased a bit, to around 250mV P-P while the collector fell by at least an order of magnitude. So please correct me if my conclusions are unwarranted, but I believe the situation is quite clear, now. The HOT has a thermal intermittent in the E-C circuit - I just can't see any other explanation for this behavior. The base signal didn't flicker, it actually +rose+ as the collector signal went away. It's as if the E-C circuit is going high-impedance while the B-C circuit remains intact. Also - if something (like C116) were shorting the C to ground, we wouldn't see a +rise+ in the base waveform.. we'd only see the base signal increase if the load on the collector were to go away - eh? IOW, It's just gotta be the HOT. From the get-go. On Thu, Nov 6, 2014 at 6:39 PM, drlegendre . <drlegendre at gmail.com> wrote:

Tony, All, On Thu, Nov 6, 2014 at 11:55 PM, tony duell <ard at p850ug1.demon.co.uk> wrote: I don't have enough experience to rate it one way or the other. All I'll say, is that (again) these readings were taken when I knew I had precious little time to take +any+ kind of measurement, before things went away - and I was neglecting / ignoring any spiky / HF components. Didn't have time to analyze the minutiae of the signal to any depth. My primary goal was to establish an easily-acquired reference to compare with any changes that might (well, that I knew would) occur. Oh, and it's like a 10MHz scope (Telequipment D54). Limited bandwidth and capability, but it does have a really excellent trigger setup.. Sure, alongside output transistors in audio amps, for example. I will. And no, nothing will really surprise me at this juncture, lol. But I do suspect that a new HOT will more than likely cure what ails it.. Time will tell. Anyone got a spare? =) Otherwise, I'll hit up eBay or whatnot.. I'll also check with the local electronics jobber, but they tend to have mad markups on this stuff. Wouldn't be surprised if they wanted $20 for a simple NTE 379.

Chuck & All, Hey, now +that's+ interesting.. I did look at the data for the MJE13007 and while I'm not exactly up-to-speed on solid state stuff, the specs seem to compare very well with the generic NTE 379 we've been discussing. And much like the NTE, suggested apps include SMPS, inverters and deflection circuits (I assume when they state the latter, "analog CRT" is implied?). Likewise, the PHE 13005 also compares fairly well, though it seems to have somewhat lower current ratings, both collector and base. The data specifies "lighting ballast" and makes no mention of SMPS or deflection service, though. I'm going to make a guess that it's not quite as fast a switching part as the others.. or maybe the 'on' resistance is a bit high-ish? But all three devices discussed have the same low beta, 8-10 min / 30-40 max. Naturally, I binned half a dozen bad CFLs only a week ago.. So Chuck, if it were yours, I take it that you'd throw a 13007 at it without a whole lot of further thought? It's looking very appealing, being I can buy 10 of them for about $4.50 including shipping, and they clearly have lots of other uses (like repairing SMPSUs and other small computer monitors). On Fri, Nov 7, 2014 at 1:30 AM, Chuck Guzis <cclist at sydex.com> wrote:

On 11/07/2014 09:41 AM, drlegendre . wrote: Well, in the first place, it is cooling--a "naked" heatsink-less TO220 isn't going to last very long dissipating 75W--simple thermodynamics tells you that. But any transistor is a compromise when it comes to geometry vs. performance and characteristics. Here's a good paper on power transistor structure that goes into pretty good detail: http://elearning.vtu.ac.in/P2/EC42/1.pdf --Chuck

I believe I said many messages ago that the horizontal output stage is a specialised SMPSU. It has been since the 1950s (yes, the 'line output valves' that I remember -- things like PL36, PL81, PL504, PL509, etc -- are choppers). In the UK the BU208 (metal TO3 package) turned up in countless horizontal output stages and SMPSUs. BUT... My experience is that some horizontal output stages (and some more conventional SMPSUs) can be remarkably fickle about what transistors will work there. And since the OP is having odd problems, I would want to reduce the number of variables. If it doesn't work with an MJE13007, then what? Is it that this transistor is not suitable, or is there some other fault? -tony

Question.. On Fri, Nov 7, 2014 at 3:18 PM, Chuck Guzis <cclist at sydex.com> wrote: Early Macs.. like Macintosh all-in-one computers? Mac RGB monitors? Some other kind of Mac? No, they just re-mark otherwise common, work-a-day parts with meaningless in-house numbers while failing to ensure that the relevant service data is freely available. ;-) Anyway, now I've half a mind to rip apart a CFL to see if it has a yummy center. What about an old 350W ATX PSU? Anything good in those?

Heh, wow.. On Fri, Nov 7, 2014 at 8:40 PM, Chuck Guzis <cclist at sydex.com> wrote: big-box home store (Menards) sells 8-packs of 60W equiv. CFLs for like $5.99 on-sale, and the sales come up quite often. Less than a buck per hole. Sales guy there once said that the local utility (here it's Xcel Energy) subsidizes the prices for some reason.. must be some idiot carbon-cutting / offsetting regulation crap, or something like that, I'd guess. But FWIW, that's what we were told. Gotta keep those polar bears in seal blubber, ya know..

"Don't be alarmed if you see smoke or fire" (coming out of your CFLs) - "It's just the end of life" http://winnipeg.ctvnews.ca/consumerwatch-cfl-bulb-safety-1.1210152 CFLs definitely do not like being enclosed or upside down; they can also create substantial RF interference.And then there's the mercury content, dim light, especially when cold, etc. But they're a good source of parts... I haven't had my LED lamps long enough yet to see how they fare in those conditions. m ----- Original Message ----- From: "Chuck Guzis" <cclist at sydex.com> To: <General at classiccmp.org>; "Discussion at classiccmp.org:On-Topic and Off-Topic Posts" <cctalk at classiccmp.org> Sent: Friday, November 07, 2014 10:09 PM Subject: Re: Osborne Vixen - Zenith 7" display t-shooting issues

Yeah, "white LED"s are pretty close to white, but only sort-of LEDs. They're fluorescent lights driven by (IIRC) ultraviolet LEDs, and they have the dimming-with-age issues most fluorescent lights do. Yeah. Up here (Ontario), some misguided fool[%] has managed to get 100W incandescents banned. (Well, not entirely. Just manufacture or import of them. Possession and use are still permitted. Also, there are some exceptions, such as rough-service bulbs.) Yeah, as near as I can figure it's a planned-obsolescence play, dressed up in "green" rhetoric, by the makers of the various short-life bulbs. I wrote about this a little on my blah; see http://ftp.rodents-montreal.org/mouse/blah/2011-11-13-1.html, the two footnotes in particular. [%] Assuming you don't push the causality back far enough to reach someone with a vested interest in this campaign.... /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

Good god! Here's a choice quote from the linked article: "The Underwriters Laboratories, the organization responsible for the safety and testing of the bulbs, said some manufacturers have changed the circuitry to reduce the negative effects when the bulbs burn out. They say people should not be alarmed if they see smoke or flames, however. ?The reality is in investigations of these incidents, we generally find it?s just the end of life,? said the organization." Not to be "alarmed" if they see smoke or flames, and "It's just the END OF LIFE"?! Wow, nice choice of words, there.. yeah, the end of life for the children sleeping upstairs, I guess. So I'm not really the conspiracy type - but either the UL has been bought-off by the CFL mfrs, or that organization is a pale, hollow shell of its former self. Since when in hell does the UL not consider +open flames+ freaked-out over excessively large or improperly placed cooling slots, that could allow a child (or idiot) to stick an object into the works of an electrical device.. and yet today they brush-off light bulbs that 'commonly' emit smoke and flame? On Fri, Nov 7, 2014 at 9:34 PM, Mike Stein <mhs.stein at gmail.com> wrote:

Not to hijack my own thread.. On Fri, Nov 7, 2014 at 11:51 PM, tony duell <ard at p850ug1.demon.co.uk> wrote: type with the 20W circular fluorescent tube. I was really surprised to find nothing but an iron-core choke in the base.. no starter device, nada.. just a choke in series with the line cord (and the lamp) I assume. It +is+ the type where you must hold down the power button for a moment to light it, so perhaps that's a (manual?) starting mechanism. But in any case, I was a little puzzled that's all there was to it, so to speak.

The normal fluorescent lamp starter is an automatic switch, effectively. Often it's a little discharge lamp (argon filled, most of the time) with bimetallic strips for electrodes. At switch on, the starter lamp strikes, the electrodes get hot, bend and touch. This energises the filaments in the fluorescent tube. Since the starter is shorted out (by the electrodes touching), it cools down, the electrodes spring apart. The sudden open circuit causes a large back emf from the ballast choke, which strikes fluorescent tube (the hot filaments in said tube emit electrons, making it a lot easier to strike). Anyway, if your magnifying lamp is anything like the one I repaired years ago, there is a momentary contact set on the on button. It's wired in the same way as the automatic starter switch in a normal fluorescent lamp. When you press and hold the on button, the filaments in the fluorescent tube warm up. When you release it, you open this circuit, causing the back emf from the ballast to strike the tube. -tony

Good News, everyone! ;-) The Osborne Vixen 7" display has been repaired.. And our friend Chuck has earned not one, but two gold stars in the process. To be fair, it's one of those situations wherein if I'd had an appropriate replacement part on-hand, it would all have been over in the first few hours.. because it turns out that the dang horiz. opt. transistor WAS the culprit all along! It had some kind of thermal fault, that made it give up the ghost as the current began to flow in earnest. Per Chuck's suggestion, I picked up (10) MJE13007 transistors from an eBay vendor - cost was less than $5 USD all-done. This was less than the cost of a single NTE 379 - the part that allegedly crossed to the mystery Zenith 1070 item. The packet arrived today, and I wasted no time installing one of the 13007s in place of the original Horizontal Output transistor - a Zenith part marked "1007". Anyway, the display came right up, looking great, and the new HOT runs very, very cool - maybe a 10 degree rise over ambient? So far, so good. A big round of thanks to Chuck, and everyone else who added some useful info to the thread. Thing is, I still have qty. 9 of these 13007 parts sitting around - would anyone like to have a spare or two on the shelf? On Sat, Nov 8, 2014 at 9:31 PM, drlegendre . <drlegendre at gmail.com> wrote:

Hi Simon, On Sun, Nov 23, 2014 at 3:41 AM, Simon Claessen <simski at dds.nl> wrote: Isn't it obnoxious when a forum member goes through all sorts of conniptions, generates dozens of exchanges - but never bothers to return and let everyone know how things came out, once they have what they want? Yeah, I dislike that kind of behavior, too. Anyway, Chuck's tip about the MJE13007A SMPS transistors serving the horizontal output role was extremely helpful. They are obviously common as a shoe and very inexpensive - less than fifty-cents apiece in lots of ten, and that's from a USA seller! If you set aside the shipping, they're more like a quarter-dollar each. I'm still new to this area, but I suspect that many small-screen monochrome displays (or even B&W TVs) might accept such a part as a replacement HOT. Heck, they might even work in a small color display, such as the 5" screen in the C-64 Executive.

----- Original Message ----- From: "Chuck Guzis" <cclist at sydex.com> Et tu, Carolus?

Me too - well, not alleys; in my case, my folks were living in a small village and the dead electronics came from the town dump. I amassed a bit of a collection of vacuum tubes, sockets, and assorted discrete parts from stuff obtained there. Aye, me too. My parents grew up during the Depression, and it shows. "Use it up, wear it out, make it do - or do without." /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

Ah... That sounds like the horizontal output stage stopped working completely (so no scan) and the thing kept working on the charge stored in the capacitors. This sounds thermal to me. Somethng is failing when hot. Maybe the HOT. Maybe a diode or capacitor in the horizontal output stage. -tony

Be careful... a low supply will lead to low EHT (HV, final anode voltage, whatever you call it) which means the electron beam is less stiff. So it will be easier to deflect. This will compensate for the low deflection amplitudes, depending on the monitor the picutre may grow, shrink, or stay about the same size. -tony