Kaypro II Video Problem
Tarkin
I recently had a Kaypro II donated to me, very compleat
with even the *dealers* manual. Recently ran it for four hours
trying to get it to talk to my PC (more on that in another post).
Screen flickered and and has now "dropped"; My prompt
"Kaypro CPM 2.2 <c/r><lf> A>", which is normally at the top of
the CRT is now approximately shifted 50% down. Sifting through my
copious documentation, the Chilton's Manual proclaims the video
board as Toshiba. Tried fiddling all the adjustments relating to
vertical: linearity,hold,&size....only made it unreadable, so
tweaked it back to normal. The internal case brace on the video
side is nylon, so I used it to variously probe, tap, and bend
the video board & tube looking for bad solder joints...to no
avail. In desperation (and casual disregard for safety), I broke
the hot-melt glue on the deflection yokes, the display rotates in
stange ways (you could say they gyre and gimbal in the wabe), but
will not deflect above the imaginary half-height line.
Does anyone have any insight into this problem?
Thanks,
Tarkin
Fred J. Scipione
"Tarkin" <nuke...@yahoo.com> wrote in message
news:1123895370....@g43g2000cwa.googlegroups.com...
Well, what you have described is a problem in the vertical
deflection circuitry on the driver board. Most likely either
one of two transistors in a 'push-pull' amplifier's power drive
stage has quit, or one of the two voltage supplies to the power
drive stage has gone. If it is in a voltage supply, it may be
a bad rectifier or bad capacitor. Bad capacitors have been
mentioned before in this group as a frequent cause in failures
when old equipment is re-powered after a long idle period.
Be warned that there are lethal voltages present in the video
circuitry. You should only attempt a repair if you are qualified,
experienced, and take proper precautions (such as having someone
with CPR training available while you work!). If your manuals
provide a schematic for the Toshiba board, use it to understand
the vertical deflection circuitry and to find key voltages you
can measure in a 'divide and conquer' search for a bad component.
If a schematic is not available, you may have to remove the
Toshiba board and create your own (partial) schematic of the
vertical section by manually tracing the connections from the
yoke connector back across the printed circuit board and its
attached components.
The good news is that your horizontal deflection circuit must
be working, and still producing (most) of the derived voltage
supplies on the board.
Video monitor repairs are like T.V. repairs (without the tuner).
Of course, T.V. repairmen are a dying breed, since the cost of
a new T.V. now equals that of most part+labor repair costs.
Still, you might find someone knowledgeable who might advise you,
or who finds a restoration project an interesting challenge.
Some books from the library on T.V. repair might be helpful too.
bud
In addition to Fred's comments, I have also been told that those symtoms
can be caused by a failed "clamping diode".
salaam,
dowcom
To e-mail me, add the character zero to "dowcom". i.e.:
dowcom(zero)(at)webtv(dot)net.
--
http://community.webtv.net/dowcom/DOWCOMSAMSTRADGUIDE
MSWindows is television,… Linux is radar.
nos...@nouce.bellatlantic.net
Fred has some good thoughts. Most likely is a failed transistor
among those.
One simple thing to check. Make sure the Video board is really
getting the power required (12V). Less than 12V with shrink the
image.
Other less likely possibility is a short in the vertical deflection
yoke.
Allison
Tarkin
This info helps me to narrow down the possible cause(es)
and potential solutions. I cruised through other posts and found out
about the excellent (Kaypro) pdf at bitsavers.org - my Dealers'
Manual is nothing more than a stripped down version.
Oddly enough, the manual instructs one to loosen the
neck clamp, reach in, and twist the yoke to square the display!
<NOTICE: What I describe next is my personal experience. Do not
attempt unless you understand there is a risk of *electrocution* and
*death* >
Reading that and having had previous HV experience,
I did try and adjust the yoke, and I already mentioned adjusting the
deflectors....I caught a poke off the flyback, 30kV is a tickle
compared to my ~60kV homemade Jacob's Ladder. The most
dangerous items on the video side of the Kaypro II:
CRT ::
Treat it like a large capacitor. The amperage is LETHAL.
Brightness Pot ::
12V hot, and ground all around Use caution.
Yoke windings & Horiz. Width ::
Both have abundant RF energy; thus RF burns are
a possiblity. Use a plastic tool when adjusting the H. Width, and
avoid grabbing the deflection coils directly....they get real warm
real quick.
The Flyback Xformer ::
There's an exposed metal bracket that seems either to carry or
induct the 30kV potential....I brushed it and lived, your mileage may
vary....
Sorry for the wordy post, but I'm trying to document this process as
I go...I googled "kaypro monitor repair" and waded through 26
pages of dead ends.....I'm hoping to save the next guy some the
hassle.
Thanks again for your insights. Testing begins sometime this
weekend, and I'll post the results.
--Tarkin
Barry Watzman
down so you can see the "raster", even where it's black.
This is really a tv (monitor) repair problem, and I'd really advise you
to seek help from someone familiar with TV or monitor repair. As
monitors go, this one is VERY, VERY simple, the entire circuit of
interest probably does not have 20 components in it. Doing anything
with the deflection yoke was probably a mistake.
primo
Speaking from experiance with hv supplies (ham radio) capacitors can
be shorted to discharge and then the next day deliver the same voltage
across the terminals. Common practice on working with hv equipment is
to short the caps to discharge them, wait a day, do it again, then
work on the equipment one handed, with the other hand in your pocket.
Picture tubes will act as a capacitor as well, with 25-40 KV
potential.
Most cheap vom meters are not rated for working with 30kv supplies, if
you want to work on it, it is best to invest in some hv probes or you
could have leakage thru the insulation of a cheap probe and take 30kv
while you probe the circuit.
The worst bite I ever took was 5kv from the supply of a geiger tube,
even that left a scar on my finger tip.
nos...@nouce.bellatlantic.net
Better is once properly discharged (through a resistor, not a short) a
SAFETY short in then installed to insure the cap can not "recharge".
>Picture tubes will act as a capacitor as well, with 25-40 KV
>potential.
Those are potentials used for the larger tubes in Color sets.
>
>Most cheap vom meters are not rated for working with 30kv supplies, if
>you want to work on it, it is best to invest in some hv probes or you
>could have leakage thru the insulation of a cheap probe and take 30kv
>while you probe the circuit.
Most are only rated to 600-1000V PEAK, Do not try to use a common
meter without the correct probe (if available) to measure HV, you
will kill the meter and it will hurt.
>The worst bite I ever took was 5kv from the supply of a geiger tube,
>even that left a scar on my finger tip.
The kaypro uses only 12-17kv for the HV. There is also a 80-130V on
the board for focus. Lethal usually not, nasty uncomfortable if hit,
definately. The biggest risk is usually getting injured when you
reflexively pull away. Myself I worry more that I'll shatter the
tube and have flying glass everywhere.
The easiest and safest is for working on the monitor is to make a lead
with a 1M ohm resistor to discharge the HV. Once discharged you leave
the lead in place to insure the anode is "safe". Do not short the HV
to ground as the momentary current surge can be very bad fr the
electronics. The 1M resistor will quickly discharge the anode but
without the high current surge. A note about the resistor, I make it
up using 5 220K ohm resistors in series and slide the whole thing
inside a plastic pen barrel for insulation.
I've fixed a lot of monitors. The HV is scary but to me implosion
still makes me work carefully. Thats a lot of glass wrapped around
a high vacuum. I've seen what happens if a tube is dropped
in a engineering lab, we ended up cleaning glass for a 30ft radius
and the gun assemble was embedded in the ceiling tile.
Be careful!
Allison
Fred J. Scipione
"Fred J. Scipione" <fjsc...@rochester.rr.com> wrote in message
news:IcgLe.14$Hx...@twister.nyroc.rr.com...
>
> "Tarkin" <nuke...@yahoo.com> wrote in message
> news:1123895370....@g43g2000cwa.googlegroups.com...
>> Screen flickered and and has now "dropped"; My prompt
>> "Kaypro CPM 2.2 <c/r><lf> A>", which is normally at the top of
>> the CRT is now approximately shifted 50% down. Sifting through my
>> stange ways (you could say they gyre and gimbal in the wabe), but
>> will not deflect above the imaginary half-height line.
> Well, what you have described is a problem in the vertical
> deflection circuitry on the driver board. Most likely either
Some further reflection on the described symptoms has given me two
more possibilities that may be likely. If the vertical driver output
is coupled to the yoke coil(s) through a capacitor, a short in that
capacitor might produce what you have described. Such a capacitor
might be mounted on the circuit board, or on the yoke assembly.
The other possibility is that a bad component in the feed-back path
from the coil drive to the vertical oscillator circuit has shifted
the DC set-point of the power stage output (and input).
bud
'Lo Allison:
Group: comp.os.cpm Date: Sun, Aug 14, 2005, 12:13pm (CDT+5) From:
nos...@nouce.bellatlantic.net
script:
>The easiest and safest is for working
>on the monitor is to make a lead with
>a 1M ohm resistor to discharge the HV.
> Once discharged you leave the lead
>in place to insure the anode is "safe".
Good tip. Thanks.
>…and the gun assemble was embedded
>in the ceiling tile.
Impressive. I can think of at least three ways that the gun assembly
could end up in the ceiling from an implosion. All three suggest that
the vacuum imparts a very high energy to the assembly.
For those who think I might be 'making fun' of Allison, I am not.
I, like many, have dropped a light bulb. (Another high vacuum vessel.)
The glass does not end up in a neat little pile below the apparent
'center of vacuum', but rather far and wide.
The vacuum will initially pull the glass in toward the center of vacuum.
The glass will pass thru the center (because the vacuum no longer
exists) and out the other side. And ricochet if it hits other objects.
Barry Watzman
> I, like many, have dropped a light bulb. (Another high vacuum vessel.)
I believe that a light bulb is filled with nitrogen. It is not a vacuum.
Terry Gulczynski
"Barry Watzman" <Watzma...@neo.rr.com> wrote in message
news:43028909...@neo.rr.com...
Argon is typical, although I suspect any inert gas will do as well.
Edison's early light bulbs did use a (near) vacuum.
Terry
nos...@nouce.bellatlantic.net
No it is a vacuum. In the case of the common lamps.
I do not know of any lighting technology that uses Nitrogen however.
Halogen and others such as mercury vapor are a very different story.
There are rare gasses Argon, Krypton, Halogens, that are often added
to enhance color, lumenence, or life. When they are added it can be
at trace amounts at near vacuum to above 1atmosphere pressure.
Allison
Tarkin
Couldn't get at testing/repair attempts 'til tonight.
The results are a mixed blessing:
The video is still knackered.
But I can replace capacitors without introducing new problems
and/or hurting myself :^)
So far I've replaced two small (47u 25V & 100u 16V) caps that
appeared to be in the vertical section- either near the wires that
come from the v. yoke or the v. signal from the board, but the tests
that I performed that made them suspect were influenced by other
components; once removed from the board, they passed.
Some reading of Sam Goldwasser's Sci.Electronics.Repair FAQ,
in the tv/monitor section, gave not only many many good safety
tips (a lot of which appeared here), but also led me believe this
is almost certainly a capacitor problem. It seems that if the rest
of the visible display is ok (i.e. has good geometry, consistent
brightness, etc), then problem is vertical retrace. What's confusing
is now I'm back to square one- is the problem on the video board,
or on the mainboard? There is vertical sync pin on the video jumper
from the mb, but removing said mb looks to be a royal pain, just to
flip it over- but I could always look at the schematic from the
bitsavers' pdf.....
Will carry on.
Should have taken Barry's advice about the raster test, will do that
next; a good raster across the whole screen would seem to indicate
a bad v. sync signal from the mb; no raster in my display-less 15%
would seem to indicate bad blanking on the video board.
Back to the labs.
Thanks to all for the tips & advice.
P.S. Nice discussion of incandescent tech!!
Terry Gulczynski
<nos...@nouce.bellatlantic.net> wrote in message
news:8175g1htvo5llrs52...@4ax.com...
This article:
http://science.howstuffworks.com/light-bulb2.htm
says the incandescent bulb (common lamp bulb) is filled with an inert
gas - usually argon. The second page discusses why an inert gas is
necessary. The article doesn't discuss the pressure in the glass
envelope, but I suspect it's somewhere between a vaccum and
atmospheric pressure.
Terry
Barry Watzman
A situation which appears like your description can occur if the display
is simply way out of vertical sync or is "synched" to a multiple or
submultiple of the correct frequency. For example, assuming that the
correct vertical retrace is 60Hz, if the display is operating at 30Hz it
could produce this, also 90Hz.
This could be an extreme centering problem caused by DC flowing through
the vertical deflection yoke. In that case, the entire raster is fine
and correct, but displaced so that the top of the raster is mid-screen.
[I wish that you had never touched the stuff on the neck of the CRT].
You need a schematic of the video board. I do agree that a capacitor is
the most likely cause. Not the only possible cause, but the most likely
cause. However, if the yoke is driven by a "push-pull" circuit of 2
transistors, then failure of one of them could do this.
A good "TV guy" could probably fix this in 20 minutes providing that you
have a schematic of the video board.
It's also possible that there is NOTHING wrong at all, just
misadjustment of some of the video board controls in the vertical circuit.
Better that you have problems with the vertical circuit than the much
more difficult to work with horizontal circuit.
Axel Berger
>No it is a vacuum. In the case of the common lamps.
It has not been so in Germany since the early twenties. America is
often said to be technically backward in some areas, but I'd be very
surprised to see you still using pre-war technology in such a
mainstream consumer area.
wild bill
>
> Halogen and others such as mercury vapor are a very different story.
>There are rare gasses Argon, Krypton, Halogens, that are often added
>to enhance color, lumenence, or life. When they are added it can be
>at trace amounts at near vacuum to above 1atmosphere pressure.
>
Then there's the arc discharge type of lamps that won't 'start'
in the dark without some loose photons ... or other gammas.
Neon is typical - ever seen a neon bulb that works okay when
the room lights are on, but flickers in the dark? Probably
fires whenever a random cosmic ray comes along (NE-2)
To assure starting, they put a little radioactive gas into them,
or a dot of material somewhere near the anode (doesn't matter
if you're running it on AC, then either element can be 'anode'
for a half-cycle anyway)
The regular sized neon bulbs we used would register on a
geiger counter. Can't remember the number. NE-51? NE-83?
whatever.
They also make thyratron-type discharge tubes (commonly
used before SCRs and Triacs came along) with some radioactive
material in them. One type of these was used to trigger nuclear
weapons. Say the magic word and catch the attention of the FBI.
Speaking of SCRs and Triacs there's some dumb woman with
a book out about Robert Noyce and The Chip. In a C-SPAN
book review program she said 4-layer devices were a dead
end not worth pursuing. What an idiot.
Bill
Norm Dresner
news:1ld6g1l25r8md4igd...@4ax.com...
Are you sure what the context of her remark was?
Norm
Tarkin
shifted down as well. Thus, the problem is with the video board.
Unfortunately, on turning the brightness back down, the pot came
apart, and I stupidly switched the unit on before
repair/replacement....
now the unit is dead. It just makes a clicking noise when switched on,
and and the 12V and 5V power section meter @ < 1V.
I am unsure as to whether or not I simply killed another component on
the video board, which is now drawing too much voltage, or if I've
screwed the p.s. up as well. As I feared from the beginning, testing
has resulted in collateral damge. I'm bidding on another unit, though
I haven't given up completely on this one. If further repair proves
futile, I'll have ample spare parts for the next one, and this case
will
likely end up holding an SBC project I'm working on. I also have
several ATX power supplies on hand, and enough knowledge to
adapt them to the kaypro, should th original p.s. be well and truly
dead.
Searches on "push pull ampliefier circuits" have shed precious little
insight on the original video problem; modern crt (tv or monitor)
circuits seem to rely on "canned" IC modules, combing a few or
more 'traditional' circuits into blocks, making this repair akin to
translating the Gallic the commentaries.....
Thank you all for your continuing input.
Barry Watzman
incandescent light bulbs.
Fred J. Scipione
> Performed the brightness test....as was suspected, the raster is
> shifted down as well. Thus, the problem is with the video board.
> Unfortunately, on turning the brightness back down, the pot came
> apart, and I stupidly switched the unit on before
> repair/replacement....
> now the unit is dead. It just makes a clicking noise when switched on,
> and and the 12V and 5V power section meter @ < 1V.
> I am unsure as to whether or not I simply killed another component on
> the video board, which is now drawing too much voltage, or if I've
> screwed the p.s. up as well. As I feared from the beginning, testing
Since the Kaypro supplies are from the days when linears ruled, you
should be able to just un-plug the connector to the video board to
see if that stops the clicking and restores the +12VDC and +5VDC at
the power-supply output. Its not like the IBM PC clone supplies
that need a proper load to make their switching regulators start-up
and keep running. I think you could also try (temporarily) running
the synch and video signals from the mother-board to a VGA monitor,
just to prove that they are working. It may not be that simple,
as you may have to look at separating the horizontal and vertical
synchs with the monitor's required polarity, driving just green or
R+G+B in parallel, etc. Still, most VGA monitors are "smart" and
may try to work with you for just a test, even if the results are
off-center or mis-colored. You sound to me like you may have the
skills to perf-board a little TTL buffer and one-shot interface as
needed.
Coming from the other end, after fixing the brightness pot you
might be able to test and trouble-shoot the video monitor separately
from the rest of the Kaypro with a 12VDC power supply and a video-
generator. The video generator does not have to be a professional
instrument; an xtal oscillator, some TTL dividers, and an AND (or
XOR) gate can give the required synchs plus a dot (or checkerboard)
video signal that will suffice for a quick check. That way you
can avoid risking the other Kaypro components. You can even use
the Kaypro II('83) schematic as a guide!
Tarkin
I keep missing the forest for the trees. I knew there was a
major difference between newer and older power supplies....the
older ones, like the ones I make, are just linear regulated with some
smoothing caps.....I just perfomed the no-vid test, drives are ok,
boots fine......now I just gotta fix the video board.
It's time to knuckle down and take another suggestion that was made
earlier....I gotta reverse-engineer the board. I'm only squemish
cuz it involves removing the HV lead from the tube; I'm not scared
of the HV, it's the amperage that the giant leyden jar I mean CRT
can hold which sets my knees to knockin. I've seen pix of the
procedure & instruments, I simply have to be absolutely certain
my waldoes are 100% insulative @ >=10KV.
Thanks again for the reminder on linear p.s.'s!!!
Tarkin
P.S. Repaired & replaced the brightness pot, it still clicks....there's
collateral damage on the board, and I'm hoping it's not the the
flyback.
Nelson Johnsrud
> Thank you-
> I keep missing the forest for the trees. I knew there was a
> major difference between newer and older power supplies....the
> older ones, like the ones I make, are just linear regulated with some
> smoothing caps.....I just perfomed the no-vid test, drives are ok,
> boots fine......now I just gotta fix the video board.
You may be interested in a similar problem I had with a TRS-80 model 4
of a similar vintage. Power supply will not start with video board
connected, but when 12V lead to video is removed, computer boots
(without video). I can then reconnect 12V video lead and all is well.
I actually added a NC momentary pushbutton in this lead to demporarily
disable it so I can boot the machine. Full thread is in comp.sys.tandy
with subject "Model 4 jump start" if you are interested.
Nels
nos...@nouce.bellatlantic.net
Well unless you heart is weak it's mostly annoying. Last time I got
zapped it was a Vt100 video board, only a bit more voltage as it's a
tiny bit larger. At least once a year I open up a PC style color
tube and repair bad connections and other simple failures
and havent been bit in a while. It's a matter of being careful
and being very deliberate. Personally the switchmode PS in
most of them with 350V loaded into a 400mF cap is scary as
that can really hurt you. There are a lot more joules in that cap!
Likelyhood of flyback getting cooked is low. On that beast it's
usually horizontal outputs or maybe vertical outputs that zorch.
Though VT100s have proven caps can fail too.
Actually if you open up the machine and spread it out so your not
working in the case it's less likely to get zapped by reaching into a
corner. The real hazard is if your "in there" and get bit you may
flinch or pull back and get cut of find something else that is live.
Allison
nos...@nouce.bellatlantic.net
wrote:
That suggests either of the video sweeps are drawing more than usual
current or the 12V PS is getting soft and can't start a full load. I
had a VT100 that had that problem. Turned out later it was the
existing PS was at it's load limit (actually it was marginally sized
for the added VT125 option and VT180 option).
Allison
Nelson Johnsrud
>>You may be interested in a similar problem I had with a TRS-80 model 4
>>of a similar vintage. Power supply will not start with video board
>>connected, but when 12V lead to video is removed, computer boots
>>(without video). I can then reconnect 12V video lead and all is well.
>>I actually added a NC momentary pushbutton in this lead to demporarily
>>disable it so I can boot the machine. Full thread is in comp.sys.tandy
>>with subject "Model 4 jump start" if you are interested.
>
>
> That suggests either of the video sweeps are drawing more than usual
> current or the 12V PS is getting soft and can't start a full load. I
> had a VT100 that had that problem. Turned out later it was the
> existing PS was at it's load limit (actually it was marginally sized
> for the added VT125 option and VT180 option).
>
> Allison
This power supply happens to be a brand new replacement that I bought a
few weeks ago. It worked fine until it had "burned in" after a few
hours of use. I changed all the e-caps on the video board (can't hurt)
with no positive results. There are no symptoms in the raster or video.
All appears normal once it starts up. I am still picking at the video
board, but your comment about a marginally sized supply hits pretty
close to the mark in this case. I have heard of others having power
supply problems with the model 4's. I think it is designed too close to
the mark, and the initial startup load is less predictable than the
running load. Part of the problem with these all-in-one units is that
the CRT filament is added to the start-up load along with the other
peripherals. The initial load of the cold filament warming up is
slightly higher than that of a fully warm filament, which complicates
the issue of startup load on the power supply. Add to that the dinosaur
disk drives which also draw a good load, and startup current can be
troublesome. I am presently going through the original power supply,
swapping all the e-caps with good quality 105 degree replacements, and I
may swap out the regulator as well. Then I will swap the original
supply back in place of the new replacement to see if there is any change.
Nels
nos...@nouce.bellatlantic.net
wrote:
>
>This power supply happens to be a brand new replacement that I bought a
>few weeks ago. It worked fine until it had "burned in" after a few
>hours of use. I changed all the e-caps on the video board (can't hurt)
>with no positive results. There are no symptoms in the raster or video.
> All appears normal once it starts up. I am still picking at the video
>board, but your comment about a marginally sized supply hits pretty
>close to the mark in this case. I have heard of others having power
>supply problems with the model 4's. I think it is designed too close to
>the mark, and the initial startup load is less predictable than the
>running load. Part of the problem with these all-in-one units is that
>the CRT filament is added to the start-up load along with the other
>peripherals. The initial load of the cold filament warming up is
>slightly higher than that of a fully warm filament, which complicates
>the issue of startup load on the power supply. Add to that the dinosaur
>disk drives which also draw a good load, and startup current can be
>troublesome. I am presently going through the original power supply,
>swapping all the e-caps with good quality 105 degree replacements, and I
>may swap out the regulator as well. Then I will swap the original
>supply back in place of the new replacement to see if there is any change.
>
>Nels
Yes, that one is marginal. Actually it's a case of QA, not all of
them fold back at the same current so in a group of systems some work
and some don't. I've seen that with many boxen out there over time.
One thing to check is if the current foldback for the 12V can be
increased somewhat. Usually it can by reducing the value of the load
sense resistor (a power resistor of some .5 ohm or less). I've seen
those over time creep up in value or were on the high side of
tolerence. Paralleling resistors 1-10 ohm across it until the PS will
start the load is one way to solve that.
You also right about filaments. The starting load can be quite hard
with them and there is significant variation from tube to tube. Added
to that manufacturing substitutions (new vendors and the like) and
design creep over time can put a solid system in the fuzzy range.
It's why vendors of quality do process testing and other margin
testing so nominal production slippage will still yeild high quality
product and low reject rates. However, even the best can screw
up and cases like the DEC VT100 monitor board with a 50V cap
in a 70V circuit that would pop a year or so later.
Allison
Tarkin
you mentioned your problem/solution, I thought instantly of the
ancient Tandon full height drives that are in my machine, competing
with the CRT for current.....
My video is back, in all it's shifted glory.
I may have tracked down the push/pull amp section - a pair of
matched transistors. A quick look-up on the NTE site shows they are
capable of 600mW continuous dissipation; I don't think I have a pair
just hanging around in the spares box, but I'll check again. Oh, and
the pnp trans is the one that is suspect....I checked the other npn's
with my meter's diode test setting; the B-C & B-E readings were
close, but the pnp (with leads reversed, of course) readings were
fairly disparate....
I may have spares after all...the 3904&3906 seem close enough in
specs; the main differences are that the 04/06's have 20V less max
C-B voltage, and their currnet-handling capability is little bit
more....
I gotta meditate a little bit more on that one before I swap 'em....
I'll at least slap my for-real transistor checker on them both, see if
the gain is w/in 10% (between the pair).
OT: the transistor is a classic too, and old Micronta....it's a cute
little
unit, served my hobby uses well.....
Oh well. Rome wasn't built in day. Fish on,
Tarkin
Barry Watzman
The Kaypro probabaly used TV scan rates, which are VERY different from
any VGA monitor, and may blow up a monitor (although some may be able to
take it). However, if you can find an old 9-pin CGA monitor, that's
another story.
Barry Watzman
screwdriver, attach a clip lead from it's shaft to ground (be sure that
the coating on the outside of the CRT glass is also grounded), and stick
it under the 2nd annode cap on the CRT (the rubber cap), touching the
contact in the center. I don't use a resistor, and I've never blown
anything, but if you want to use a resistor, fine. I would start with
50K to 100K or so (not megohms), but after you do it with the resistor,
then do it without the resistor, and hold it on the 2nd annode contact
for a few seconds.
Barry Watzman
On many if not most monitors, the CRT filament is powered from a
"secondary" of the flyback transformer. The "secondary" is often
nothing more than 1 to 3 turns of wire.
nos...@nouce.bellatlantic.net
>I may have tracked down the push/pull amp section - a pair of
>matched transistors. A quick look-up on the NTE site shows they are
>capable of 600mW continuous dissipation; I don't think I have a pair
>just hanging around in the spares box, but I'll check again. Oh, and
>the pnp trans is the one that is suspect....I checked the other npn's
>with my meter's diode test setting; the B-C & B-E readings were
>close, but the pnp (with leads reversed, of course) readings were
>fairly disparate....
>I may have spares after all...the 3904&3906 seem close enough in
>specs; the main differences are that the 04/06's have 20V less max
>C-B voltage, and their currnet-handling capability is little bit
>more....
>I gotta meditate a little bit more on that one before I swap 'em....
>I'll at least slap my for-real transistor checker on them both, see if
>the gain is w/in 10% (between the pair).
Dont worry the gain spread. However the lower Vcbo is a concern
as you need enough to stand the voltlage in that part of the circuit
and any runing or other inductive effects that may create voltages
that exceed the devices. However since I buy them by the pound
I'd give them a shot as the worst that can happen is they fail.
Oh, a lamp (try 1157 tailight lamp) in series with the 12V is a great
current limiter and short circuit protector. An old trick I've used
for audio and other amps where a short can be destructive
if there is no current limiting.
Allison
Tarkin
On a hunch, I retrieved what I believed to be OEM specs on the
transistors....both the suspect and replacement. NTE only provides
specs for their replacement; the specs on the actuals were a *lot*
better matched. My spare ended up being a PN2907A from Nat.
Semi. This transistor could handle ~6X greater current, and 10V
more C-B & C-E voltage; nevertheless, the gain was exactly the
same, so as long as the current going in doesn't overshoot, I'm ok.
Long story short, it was a drop in replacement. Video came up, in
all it's shifted glory. The original transistor checked out ok out of
circuit. %^&*!!.
Rethinking everything leads me to believe that the problem is in or
near the yoke coils and associated components. There are two
resistors, one on each coil. Leads from each side of the resistor/
coil go to the video board. I'm beginning to suspect the larger
output caps at this point, or those resistors. The only reason I
haven't
replaced the big caps yet is their oddball values....6.8u @ 160V,
1000u @ 100V, etc.....I don't have those in my spares, I'll have to
track 'em down from Digi-Key or other some such outfit, which
means finding something else to order to justify the shipping.
I've done some other wacky things, which just put noise on the HO...
funny looking, but ineffective. Suspecting that perhaps the PS wasn't
delivering a stong 12V, I hooked up one of my homebrew supplies
in parallel: When I dropped the V on the homebrew, the display
shifted down further; when I cranked it full bore (~16.5V), the display
still wouldn't go above what I've come to call the 'ceiling'. The home-
brew didn't have the strength to run the CRT by itself. I'm still
debating on whether or not to try feeding the CRT from my spare
ATX PS. Anywho, it was watching the display not creep above the
'ceiling' that reinforces the suspicion about the coils and assoc.
components.
For the curious, the transistor I replaced was marked A1015-Y.
Thank you all for your continued input. I'd be bald now without your
help!
Tarkin
glen herrmannsfeldt
(snip)
> I, like many, have dropped a light bulb. (Another high vacuum vessel.)
> The glass does not end up in a neat little pile below the apparent
> 'center of vacuum', but rather far and wide.
Low wattage light bulbs are vacuum filled, higher are inert
gas, usually nitrogen.
There is a balance between the evaporation rate of tungsten
and the heat (power) loss due to the gas. It also changes
with the lamp voltage, I believe lower voltage is a little
better (more efficient overall).
-- glen
Fred J. Scipione
> More news from the trenches.
> Rethinking everything leads me to believe that the problem is in or
> near the yoke coils and associated components. There are two
> resistors, one on each coil. Leads from each side of the resistor/
> coil go to the video board. I'm beginning to suspect the larger
> output caps at this point, or those resistors. The only reason I
> haven't
> replaced the big caps yet is their oddball values....6.8u @ 160V,
> 1000u @ 100V, etc.....I don't have those in my spares, I'll have to
> track 'em down from Digi-Key or other some such outfit, which
> means finding something else to order to justify the shipping.
Just out of curiosity, what is the resistance of (each of) the yoke
coils and resistors. I have this theory that the drive transistor
dissipation rating sets them somewhere in the 8 to 80 ohm range for
drive by the 12VDC supply. Larger values would indicate a higher
voltage secondary supply is used by the vertical output driver
stage. Some where around 24 ohms could work with the 1,000 uF
output coupling cap you just mentioned.
glen herrmannsfeldt
(snip)
> Then there's the arc discharge type of lamps that won't 'start'
> in the dark without some loose photons ... or other gammas.
To get closer to the subject, I knew someone with an EPROM that
would only work in the light. The problem was that it was part
of a camera controller and was supposed to be used in the dark.
It is also hard to find the problem, as one likes to debug with
the lights on.
-- glen
glen herrmannsfeldt
(snip)
> I do not know of any lighting technology that uses Nitrogen however.
Higher wattage, maybe 60W or more, incandescent lamps are gas filled.
It might be that argon is enough better and not much more expensive
than nitrogen to be used.
The main limitation on incandescent lamps is evaporation of the
tungsten. As the temperature increases so does the evaporation
rate. The coiled or coiled coil filament allows for a gas filled lamp
without too much heat loss to the gas. Collisions with gas molecules
reduce the evaporation rate as there is a reasonable chance for the
tungsten atom to come back to the filament.
> Halogen and others such as mercury vapor are a very different story.
> There are rare gasses Argon, Krypton, Halogens, that are often added
> to enhance color, lumenence, or life. When they are added it can be
> at trace amounts at near vacuum to above 1atmosphere pressure.
Halogen lamps are incandescent lamps that are halogen filled, usually
iodine, and are designed such that the bulb wall can reach a fairly high
temperature. Evaporated tungsten deposits on the wall. At a high
enough temperature it will react with iodine to form tungsten iodide
gas. At the filament temperature tungsten iodide will decompose
redepositing the tungsten back on the filament.
This allows the filament to operate at a higher temperature without
an excessively short life, and so increase the efficiency of the lamp.
-- glen
Tarkin
the drives) weighs in at 0.8 - 1.0 ohms, had no resistor, and
had it's solder joints wound with what appeared to be medical
tape. The one on the right weighs in at 4.6~5..2 ohms, has
a parallel 22K 10% tolerance resistor. Some quick calculation
shows that the resistance should be in the .79~.99, if the yoke
coils are approximately equal. The solder joints on the right
look pretty bad; I will disconnect the resistor during the repair
and ohm it out, and the coil as well. The coils themselves are
woind on plastic forms, and are fitted on the neck in over&
under configuration. I'll post the results as soon as is
practical.
Thanks,
Tarkin
Mark Kent
glen herrmannsfeldt <g...@ugcs.caltech.edu> espoused:
> bud wrote:
>
> (snip)
>
>> I, like many, have dropped a light bulb. (Another high vacuum vessel.)
>> The glass does not end up in a neat little pile below the apparent
>> 'center of vacuum', but rather far and wide.
>
> Low wattage light bulbs are vacuum filled, higher are inert
> gas, usually nitrogen.
Shouldn't that really be 'emptied' rather than 'filled' :-) at
least for the vacuum, anyway...
>
> There is a balance between the evaporation rate of tungsten
> and the heat (power) loss due to the gas. It also changes
> with the lamp voltage, I believe lower voltage is a little
> better (more efficient overall).
>
> -- glen
>
--
end
| Mark Kent -- mark at ellandroad dot demon dot co dot uk |
Truly simple systems... require infinite testing.
-- Norman Augustine
glen herrmannsfeldt
(snip, I wrote)
>>Low wattage light bulbs are vacuum filled, higher are inert
>>gas, usually nitrogen.
> Shouldn't that really be 'emptied' rather than 'filled' :-) at
> least for the vacuum, anyway...
There was recently a story about some place that was so cold it
was minus 50 degrees below zero.
Is half as fast the same as twice as slow?
Is half as slow the same as twice as fast?
I think, then, that removing the air is the same as supplying
the vacuum. The vacuum is lost, removed, gone, when air
leaks in.
-- glen
bud
'Lo Glen:
Group: comp.os.cpm Date: Tue, Aug 23, 2005, 6:17pm (CDT-2) From:
g...@ugcs.caltech.edu (glen herrmannsfeldt)
script:
>Low wattage light bulbs are vacuum
>filled, higher are inert gas, usually
>nitrogen.
So far, that's two for nitrogen,… one for Argon.
I would have thought that the nitrogen would convert the tungsten to
tungsten nitride, making it more brittle.
So much to learn.
I have been told that some of Edison's original carbon filiment vacuum
lamps are still burning. (Probably at lower voltage.)
Far as I know, 'vacuum tubes' still have vacuum. (With the exception of
the occasional special types that need a particular gas.)
salaam,
dowcom
To e-mail me, add the character zero to "dowcom". i.e.:
dowcom(zero)(at)webtv(dot)net.
--
http://community.webtv.net/dowcom/DOWCOMSAMSTRADGUIDE
MSWindows is television,… Linux is radar.
Fred J. Scipione
> Looking at the unit from behind, the one on the left (towards
> the drives) weighs in at 0.8 - 1.0 ohms, had no resistor, and
> had it's solder joints wound with what appeared to be medical
> a parallel 22K 10% tolerance resistor. Some quick calculation
> shows that the resistance should be in the .79~.99, if the yoke
> coils are approximately equal. The solder joints on the right
> look pretty bad; I will disconnect the resistor during the repair
> and ohm it out, and the coil as well. The coils themselves are
> under configuration. I'll post the results as soon as is
> practical.
>
> Thanks,
> Tarkin
Although I can imagine that the vertical and horizontal yokes
could be wound differently, I know of no reason why the two halves
of either yoke pair would differ as you have described. Two 5 ohm
coils in series for the horizontal yoke assembly sounds right to
me. I also can't see what the 22k ohm resistor could do at the
60 Hz vertical frequency, but it could adjust horizontal linearity
against a high reactance at the 15.75 kHz horizontal frequency.
On my Kaypro the yoke leads have a 4-pin connector. The connector
can be unplugged, to allow easy measurement of the yoke assemblies
and of the yoke drive output circuitry.
Tarkin
back up for another round of debug, powered up, and.....
Garbage. Shifted garbabe. Reset did nil. Powered down for ten
seconds, back up, no sound sound from the drives. Something,
either in the controller circuitry or the drives themselves is dead.
12V is good, 5V is good, I'm very bad.
The silver lining is I'm due to pick up another Kaypro this Friday.
The dead unit will serve as a source for spare parts, and I may do
something with the case.
Thank you all for your help and input. Should anything happen to
the replacement unit, I have a much better grasp on what to look for
and where to start debugging.
Thank you all again,
Tarkin.
P.S.
Fred - I completely neglected to describe the yoke connector. I
disconnected and ohmed it out- the red & blue connections, on
the left side, read .5 ohms, and the yellow & green (which had the
22K resitor in parallel), read 4.9 . If the 22K resistor was some kind
of horiz. hack, that leaves the other connections as the vertical coil,
and the .5 ohm would seem to indicate a short in the coil windings,
as was suggested earlier in the post.
Lee Hart
> It's time to knuckle down and take another suggestion that was made
> cuz it involves removing the HV lead from the tube; I'm not scared
> of the HV, it's the amperage that the giant leyden jar I mean CRT
> can hold which sets my knees to knockin.
I fixed TVs for a living for over 10 years. To discharge the HV lead on
the side of the CRT, connect one end of a clip lead to ground. Connect
the other end of the clip lead to a small flat-blade screwdriver. Slide
the screwdriver under the rubber cap to touch the metal clip inside.
There will usually just be one "pop" as it discharges. I usually leave
the screwdriver and clip lead in place until I'm ready to power up
again, as the capacitance of the CRT takes a little while to fully
discharge.
The Kaypro video monitor is actually a pretty generic item. They bought
them from Ball Bros, Motorola, and others. It will probably be easier to
find a complete 9" monitor from some other dead computer and install it
than to mess about trying to fix the one you've got.
--
"One doesn't discover new lands without consenting to lose sight of the
shore for a very long time." -- Andre Gide
--
Lee A. Hart, 814 8th Ave N, Sartell MN 56377, leeahart_at_earthlink.net
Axel Berger
>higher are inert gas, usually nitrogen.
As far as I know they are not and they can't be. The filaments are
heated to very near to their melting points and nitrogen is a reagent,
it is used in case hardening for iron alloys. I do not know a lot about
the chemistry of tungsten and osmium but would be very surprised if at
those temperatures they did not react and make nitrides.
The fillings are noble gases, mostly argon and some xenon.