|
Vintage Computers Any vintage computer systems, calculators, video games etc., but with an emphasis on 1980s and earlier equipment. |
|
Thread Tools |
3rd Feb 2019, 11:20 pm | #1 |
Retired Dormant Member
Join Date: Feb 2019
Location: Northumberland, UK.
Posts: 4
|
Microvitec Cub monitor model identification and repair
Hello,
I'm new here and this is a place of gurus. Forgive me for troubling you. I have a Microvitec Cub monitor, acquired several years ago. The Microvitec label on the back has a serial number but where the model number should be, it's blank. If it helps, there's also a 'National Power' stock label on it. A hand-written sticky label on the side says "1441MS4". The obvious answer is that it's a 1441 M series 4. It's definitely an M series but I can't be certain whether it's a 1441 or a 1431 and whether it's a series 3 or 4. From circuit diagrams obtained for the series 3 and 4, it looks more like a series 3 but I'm uncertain and likely need to look harder. Did the tripler change from the series 3 to the 4 and is this the easiest way to tell the difference? How does the 1441 differ from the 1431 internally? Any help would be gratefully received. Apart from a Microvitec label on the edge of the main board, there is no other labelling to help. From the series 3 circuit diagrams, the tube base board appears to match the hires rather than standard res diagram and so my inexpert conclusion is that it's a 1441. It worked nicely for several minutes when it arrived and then displayed a single thin line in the middle of the screen without even a phut to accompany the change. 2 years later, I've finally got around to thinking about it. I've been looking at the circuit diagrams and trying to determine the cause of the frame collapse and I've been looking in the area of the frame coils circuits, fusible resistors, electrolytic capacitors and the vertical control IC, IC301, a TDA1170s. No damaged components leap out on a visual inspection but what does that really tell me as there could be dry joints and cracked traces: 1983 was a long time ago and no doubt it was stored in a loft for a long time before it travelled in the back of a courier's van to me. I'm trying to drag my electronics knowledge back through a 25 year void and it wasn't the greatest then anyway: much textbook study but very basic practical and little in the way of diagnostic repair. Any help in guiding my thinking through this would be great. I'd like to work it out rather than throw components at it and the hope is to understand it, repair it and then use it with the BBC Micro that I've fettled. The problem has already gotten under my skin. I've been looking at building a safe discharging method for the capacitors and CRT. Buying resistors of sufficient wattage to cope is proving more of a challenge than I'd expected but I clearly don't know enough. Any help there would also be appreciated. I was concerned that the instantaneous wattage of the CRT discharge through 10 M Ohm would be initially in the kWs for a fraction of a second before diminishing. I've attached a top and back view of the monitor's internals. The tube base board is loose and the main board is off the stand-offs with most connections removed as I had dismantled it to look under the main pcb - just to explain the photos. https://jmp.sh/TC8HQ1U https://jmp.sh/D1eECTb |
4th Feb 2019, 1:32 pm | #2 |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Microvitec Cub monitor model identification and repair
Welcome to the forum
Series 4 didn't have a tripler, so that looks like an earlier model. I'd say Series 3, but will also add that I've no idea about Series 1 or 2. Did they exist? Perhaps they did, but pre-dated the Cub? I must say that I'm no expert on these monitors specifically, but we've discussed them many times before, and what we've all learnt from these conversations is that there were many, many variations made over the years. Sometimes model numbers (when present) don't always make sense. Perhaps innards got swapped around to keep them alive - they were bought in large quantities by many organisations - or perhaps different innards were installed from new to keep them flying out the factory doors? Certainly, I have one at home which I know is original - the chassis is definitely Series 3, but the model number implies Series 4! My ones have a beige PCB with a black silkscreen that includes more clues about the series - if your board has similar markings, they'll be at the back, towards the middle. I think I've seen high-res tube bases in standard-res monitors. Maybe they decided that having 2 separate assemblies was more costly than have just one (even if it was perhaps slightly more expensive?). Or again, maybe it was just about keeping the production lines going? Broadly, the high-res tubes had a dark background, and the standard-res tubes were light grey. I say "broadly" because there are many variants, and as many pedants waiting to correct me. If in doubt, look up the datasheet for the CRT fitted. The best ones had a dot pitch of 0.31mm, which is very respectable for a BBC micro. I suspect those CRTs were intended to VGA monitors back in the day... OT: I did once make a lovely BBC Micro monitor from an CGA/EGA monitor - I just had to build a simple sync separator. It was the clearest, sharpest display I think I've ever seen on one of these. Unfortunately, I couldn't bear the line whistle, and while I managed to damp down some of the inductors, I couldn't eliminate it completely. That was 20 years ago - I wish I'd kept it, as I wouldn't be bothered by it now! I can still hear 15.625kHz with no problem, but I'm definitely much less sensitive than I once was Anyway, I agree that dry joints is the most likely thing, but it is very possible that components have failed with no visible signs, unfortunately. Most people discharge the CRT by simply shorting it. Years ago I made an insulated probe that incorporated a 1k wire-wound resistor that made the crack a bit less scary - more "because I could" rather than any actual concrete reasons. Whatever method you use, it's important to understand that a CRT will recharge itself if left unconnected (read up on "dielectric absorption"), so leave a croc-clip lead attached while it's being worked on. When you discharge, make sure you're discharging to the outer "dag" via the braid rather than the monitor's chassis. It is possible to work with a charged EHT providing you do not disconnect the earthing between PCBs and dag. However, this is risky, so I mention it for completeness only - not recommended if you're not sure about what you're doing. Cheers, Mark |
4th Feb 2019, 5:30 pm | #3 |
Rest in Peace
Join Date: Apr 2008
Location: London, UK.
Posts: 2,508
|
Re: Microvitec Cub monitor model identification and repair
A prime suspect for frame collapse is the reservoir capacitor for the frame output flyback supply, possibly 1000uF/40V or similar and which might be that orange ROE cap visible near the edge of the board in the pic. Even 25 years ago they were failing so regularly that I used to change them on sight, working or not. I can't recall if it takes out a fusible resistor with it. The frame output IC and other stuff did fail too but that was the most common cause.
|
5th Feb 2019, 8:55 pm | #4 |
Retired Dormant Member
Join Date: Feb 2019
Location: Northumberland, UK.
Posts: 4
|
Re: Microvitec Cub monitor model identification and repair
Thanks to both of you for swift and very helpful replies.
I've made a search for the tube data but can't find anything. The label states, "NEC 370KTB22 - TC09, No. 30704288". The white plastic surround at the end of the neck tube has "JVC" embossed on it. The ROE 1000uF 40V capacitor has failed with no visible signs. Given what was said about having to change these often, does the circuit design lead to frequent failures of this capacitor? Should I stock up? I appreciate that a loft stored monitor from 1983 powered up in 2019 will present very different failures from its day. Fusible resistors aren't something that I've seen before. Is it possible that they can fuse without any visible sign? I'm currently determining which are fusible from the circuit diagram and parts list. |
5th Feb 2019, 10:42 pm | #5 |
Rest in Peace
Join Date: Apr 2008
Location: London, UK.
Posts: 2,508
|
Re: Microvitec Cub monitor model identification and repair
You shouldn't need to change that cap again - the originals were just unreliable. I don't have a manual but follow the power rail back from that reservoir cap towards the LOPT, and you should find a diode and the fusible resistor. They can indeed fail without visible signs of heating. The diode might be bad instead / as well. Before changing, check that a short to ground is not still present on the rail due to a failed frame output IC.
|
6th Feb 2019, 12:43 am | #6 |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Microvitec Cub monitor model identification and repair
I've had a quick look in the manual, and the resistor you need to check is R235. It should be 10 ohms. Perhaps this would be a suitable replacement: https://cpc.farnell.com/neohm-te-con...-5w/dp/RE07246
I also looked for the CRT details, but unless I've missed it, the manual doesn't seem to give details about CRT types. Certainly, 370KTB22 is the model number (370mm meaning 14", of course), but I also failed to find anything useful about it online. I'm afraid my examples are up in the attic, so it won't be easy to check the part numbers of those. But as I say, look at how dark the screen looks. The hi-res ones are almost black - lovely |
6th Feb 2019, 8:52 am | #7 | |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Microvitec Cub monitor model identification and repair
Quote:
The common faults for zero frame deflection current in the yoke are failed power supply to your frame oscillator or frame output stage, all of that is in your TDA1170 IC (perhaps a fusible resistor powering it is open). The TDA1170 itself could have failed, they can run hot. Even if the power supply filter capacitor on the TDA1170's power input supply had failed unlikely it would stop the IC from working completely and there would be some frame deflection. (The frame output stage is very much like its power audio output stage counterpart, its a linear amplifier primarily. The frame deflection output amplifier (in your IC) is designed to apply a trapezoidal shaped deflection voltage to the yoke, which produces a linearly increasing ramp of current with a rapid crt beam flyback, back to the start of vertical scan) So to fix this fault the best move is to power the monitor with the brightness set to near zero (you don't wan't to burn a horizontal line into the screen) then test the power supply to the TDA 1170. If the supply is there, I would suspect the IC, or one of the components immediately associated with it. If its power supply feed has failed, find the reason why. If you can't get the exact schematic for your particular monitor, just look up the data sheet for the TDA1170 instead, the circuit will be nearly identical to the example circuit there. Here is a useful data sheet: https://datasheetspdf.com/pdf-file/1.../SGS/TDA1170/1 |
|
6th Feb 2019, 2:14 pm | #8 | |
Rest in Peace
Join Date: Apr 2008
Location: London, UK.
Posts: 2,508
|
Re: Microvitec Cub monitor model identification and repair
Quote:
The OP mentions his capacitor had also failed - I would have been pretty surprised if he had said it was OK! I expect R235 is O/C. |
|
6th Feb 2019, 3:36 pm | #9 |
Retired Dormant Member
Join Date: Feb 2019
Location: Northumberland, UK.
Posts: 4
|
Re: Microvitec Cub monitor model identification and repair
Thanks again to all for your help. The circuit diagram had pointed me to R235 and possibly D201, but confirmation of those helped enormously.
R235 has gone open circuit. The accompanying diode, D201, appears to function correctly but that conclusion comes only though multimeter resistance and diode checking. The plan is to replace the faulty components and then check around the TDA1170 IC. The tube is black, as far as I can determine, and the main board includes C225 which is annotated as only being fitted to med and hi res models. |
6th Feb 2019, 10:23 pm | #10 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Microvitec Cub monitor model identification and repair
That sounds nasty. I have never seen a ROE capacitor. Though I once had a ERO brand electrolytic short out. I disassembled it to find out why, it looked like there was probably a piece of metallic particulate matter embedded in the insulation between the foils, probably contamination at the factory.
|
6th Feb 2019, 11:33 pm | #11 |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Microvitec Cub monitor model identification and repair
Sounds like you have a nice one
Sadly, a lot of the Cubs where I work were low-res. They were bought to be used as composite monitors in edit suits and similar, so they had the PAL decoder option. Much cheaper than "proper" broadcast monitors, and in a nice, convenient cube shape, of course - but the picture quality was nowhere near as good. But for where they were used, it didn't really matter - it was a fairly pragmatic decision for an organisation (or at least, a part of the organisation) that has nothing like the amount of cash than some like to believe. While looking at the diagram for C225, I was able to confirm something that I was fairly sure about when I replied earlier. As I didn't have the diagrams to hand, I decided it was safest to not mention it. However, here goes: Inside the tripler is a chain of resistors - including the focus pot. These resistors will discharge the CRT. I don't know how long it takes, but certainly that CRT will hold its charge for a matter of minutes rather than weeks. When I eventually get around to sorting out my two examples, I will test for this and report back. In fact, given the amount of questions about these, plus the rather scattered nature of the information available out there, I'll probably put some effort into making a feature on my website about them. At the same time, I'll scan the diagrams. One day... Something else I double-checked: As is usually the case, the dag is not connected directly to chassis - it goes via a resistor, and this is used to sense the EHT current and control the beam limiting circuitry. So my earlier advice to discharge to the dag, not the chassis, was good advice. OK, it's standard practice, but occasionally you meet sets where the dag is connected directly to ground. Oh, and 25kV into 10Mohm is 62.5 watts. I didn't give it a thought when I read your first post the other day, but upon re-reading it, it caught my eye - perhaps as a result of thinking about the resistors fitted inside the tripler (which will be higher than 10M, but won't be G-ohms). If there was 100M total in series, then that's about 6 watts - my guess is they'll be perhaps double that? Something else to attempt to measure when I get them on the bench. I'm also tempted to measure the capacitance of the CRT - those two measurements will enable me to figure out how quickly the EHT discharges. So many thoughts, such little time. Forgive the ramble! |
7th Feb 2019, 11:12 pm | #12 |
Retired Dormant Member
Join Date: Feb 2019
Location: Northumberland, UK.
Posts: 4
|
Re: Microvitec Cub monitor model identification and repair
Thanks again for all of the very friendly and generous help. This thread is a great read and looking around at other threads, the knowledge herein is remarkable. My interest lies in 80s computers and their various monitors and there is so much to read and learn here.
I've been looking at suitable replacements for the ROE 1000uF 40V capacitor and hope that I can ask another question. Would a Panasonic 1000uF 50V +/- 20%, such as: https://cpc.farnell.com/panasonic-el...7437?st=1000uF be suitable, or am I on the wrong track and missing something from the specification? If I can't find a 40V capacitor, is it acceptable to go for the next higher voltage? Is there any problem with doing this? Is there a sensible limit of difference in voltage when replacing capacitors? The parts list states: "C224 1000uF 40V ALUM ELECT CAPACITOR - 20+50% RADIAL" Is the " - 20+50%" a tolerance of -20% and +50%? Hence is +/- 20% the most appropriate as a replacement? I was considering replacing all of the electrolytic capacitors on the main board. Is this wise? Is there anything else that I should be looking at replacing, in preventative paranoia, or that will improve things: assuming that the replacement of the known faulty capacitor, fusible resistor and diode leads to success? |
7th Feb 2019, 11:26 pm | #13 | |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Microvitec Cub monitor model identification and repair
Quote:
Also here are my thoughts on discharging the CRT: There is really no need to discharge it most of the time. The charge is built up in the glass bulb between the internal and external aquadag, the internal dag is connected to the final anode clip. The internal aquadag is also connected to the final anode structure in the CRT's gun assembly, but that is insulated (by a vacuum) from the pins that connect to the tube socket and the other electron gun metallic elements that are connected to the tube base. I have never received a discharge of any kind unplugging the CRT base and touching its pins, even when the bulb remains charged. After all, if there was any leakage across the gun's final anode to focus electrode or other gun structures, the bulb charge would already have dissipated. So with the set switched off, you can take off the CRT socket and pcb assembly off the CRT without being exposed to the charge stored in the CRT's bulb. You only require to discharge the bulb if you plan to take off the anode connection (EHT clip) which for many fault finding exercises is not required. There is only one common reason to discharge the CRT, that is if you are removing the final anode cap and might touch that lifting out the CRT or carrying the CRT across a room and get a zap and drop the CRT on the floor as a result. The total charge stored in the CRT could give an an unpleasant zap, but not too risky, but it might make a person drop the CRT. Some people short the anode cap out to the external aquadag or ground, it is not really a great idea actually, without current limiting, as it is possible the dag on the internal connection to the anode cap could be damaged by very high peak currents. It would be better to do it with a 100M Ohm high voltage rated resistor, if you really want to do it. Therefore, the best tool to do it with, is a standard EHT voltage probe that has this resistor already built into its body and you can slip its tip under the plastic EHT cap and its flying wire is grounded to the sets chassis or external CRT aquadag. So with the set switched on you measure the EHT voltage, with the set off, the probe gently discharges the CRT. Using wires, screwdrivers etc etc, you might accidentally discharge it to a pcb component and create a new fault. |
|
8th Feb 2019, 12:15 am | #14 |
Dekatron
Join Date: Jun 2009
Location: Staffordshire Moorlands, UK.
Posts: 5,263
|
Re: Microvitec Cub monitor model identification and repair
ROE and ERO capacitors (Roederstein) are generally pretty well regarded in industry but in the cub they seemed to be a failure point and when they've gone they seem to be rather old already. I know the cubs at our school (showing my age) were switched on first thing in the morning and only turned off at night by a key operated contactor on the wall. Most of those yellowy gold caps seem to go short, ISTR when you took them off the board they were disconcertingly lightweight. Any cub these days probably needs all the electrolytics checking by now, a good resolder of the power resistors, smpsu and lop transformers etc, but we've been here in another thread.
__________________
Kevin |
8th Feb 2019, 1:31 am | #15 | ||||
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Microvitec Cub monitor model identification and repair
Quote:
If you look at the Microvitec schematic, it's trivial to see how the potential of the dag, relative to chassis, is used to effect the beam limiting. In many other sets, the same signal is used to vary width and height to reduce "breathing". If you'd like more details, there are countless explanations of this in manufacturers' manuals and in articles in Television magazine. Of course, there will be variations on the theme, but this is how it's usually done, I promise The sets that did have earthed dags (often Sony) took the negative output of the EHT system to ground via a resistor. The effect is the same, give or take the time constant formed by the capacitance of the CRT. The Sony schematics can be a pain to follow, especially on a screen, but usually the width and height control elements can be fitted on the screen together with the line output stage, so it's easy to see how they work, and the signal from this that goes off to the signal processing parts of the set is usually labelled "ABL" (auto beam limiting) - knowing that helps you to follow a Sony schematic. Quote:
However, what you didn't say: you must keep the lead from the dag connected to the chassis (usually connects to the tube base PCB at least - for flashover protection - but there might be other connections in some cases). If you don't do this, you have a 25kV battery connected to the chassis via the EHT cable, meaning there is 25kV between the chassis's PCB and the dag. You usually get away with this because the EHT generator usually contains no reverse path, but in this case, there is a chain of resistors joining EHT to chassis, and so you could end up with -25kV at the dag, relative to the chassis. It won't kill you - especially as it comes from a high source impedance - but there's a risk of damaging static-sensitive components, etc. Far easier to just discharge the CRT to eliminate all doubt IMHO. * Be aware that some Sony CRTs contain internal resistors from EHT to various connections on the tube base pins. See attached... Quote:
I also mentioned the dielectric absorption point - easily overlooked. Quote:
Hope this helps, Mark |
||||
8th Feb 2019, 1:45 am | #16 | |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Microvitec Cub monitor model identification and repair
Quote:
I haven't seen it for a while - really must look for it. |
|
8th Feb 2019, 1:48 am | #17 | |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Microvitec Cub monitor model identification and repair
Quote:
That , makes sense if the negative of the EHT generator is connected to the external dag, then through the resistor to ground the beam current could be sensed there. I have never owned a set that did this, the only ones I have seen have the external dag grounded, but mostly the ones I worked on were early models from the 1970's and early 80's.Though they might have has current sense resistors in the EHT negative. Also, I guess for the CRT types where there is a resistive connection between the internal dag and the gun electrode structures, the stored bulb charge should self dissipate as there is a complete discharge pathway with the power off. PS: What is a typical value current sense resistor for the application, also are they a physically large part ? Last edited by Argus25; 8th Feb 2019 at 1:57 am. |
|
8th Feb 2019, 3:42 am | #18 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Microvitec Cub monitor model identification and repair
I just had a look at the last color monitor I worked on, the IBM5153, it has a current sensing resistor network in the ground return of the EHT supply, which they use to modify the CRT grid 1 voltage for brightness compensation, but the aquadag is grounded (not returned to the EHT negative).
Also, of note, there would be no need in this monitor to discharge the CRT, because the entire EHT supply has a voltage divider across it that supplies the focus pot and G2 voltage.So it self discharges. |
8th Feb 2019, 3:04 pm | #19 | ||
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Microvitec Cub monitor model identification and repair
Quote:
As I say, I much prefer to not take chances - it's practically zero effort to discharge (then strap a lead there to stop it charging up again), so why waste time analysing it when there's probably more pressing matters to think about? Quote:
As to the size of them, see earlier comment: they're only handling a few milliamps... |
||