[Mirrors]
Notes on the Troubleshooting and Repair of Video Cassette Recorders
Contents:
It would be nice if it were possible to output still frames from a PC,
for example, to record computer animation on video tape. This would permit
images to be generated slowly and then played back in real-time.
However, there are a couple of problems with attempting to cleaning record
single frames on a consumer grade VCR:
* Without moving the tape, only a single field (of the two interlaced
fields in a video frame) can be recorded since the tracks for the A and
B heads will be superimposed. I doubt that any VCR not specifically
designed for single frame recording has any support for moving the tape
in this manner.
* The control and synchronization circuitry to cleanly switch the
record for a single frame may not exist. This will depend on the
model - the more sophisticated the editing functions that are supported,
the more likely that this precision will be supported.
* The VCR must have a flying erase head or you must use new or pre-erased
tapes to avoid the rainbow interference on the first few hundred frames
of any recording made over old video.
Other than that, there is no reason that the video writing circuitry
cannot be turned on during pause - some VCRs will do this if you
go into record mode while in pause.
Obviously, anything you can do from the front panel or remote you can
do under computer control. There could be hidden functions accessible
via a special connector or key sequence but you would need documentation
for your unit which may not be readily available, if at all.
Here is one approach to using a PC to program multiple VCRs. Obviously,
the techniques described below can be extended to more complex functions.
Feedback could be added to inform the PC of end-of-tape or other fault
conditions.
(From: Bill Mohler (bill@cs.oberlin.edu)).
We did a project to control multiple VHS VCRs where time and cost
were major factors. Our VCR's were the same brand (assorted models with
same IR codes), so we hacked a remote to interface to a PC's parallel port.
The basic idea was to have the PC select a VCR then "push" a button for
function.
The remote hack was simply an analog switch (TTL input) across the switch
contacts and a 754XX peripheral driver to select an IR LED mounted right
in front of each VCR's IR window.
The software was simple. We only needed 4 VCR's and 4 functions, so we
split the 8-bit printer output into two 4-bit commands ("VCR select" and
"function") and "poked" away. Not bad for a days work.
You could use decoders or the printer control port to get the extra "bits"
you need.
Some VCRs and TVs may have a selector switch or be universal but you would
have to check the manual.
Power wise using a transformer will probably be fine. The power line
frequency is not used for anything in the TV or VCR except possibly the
clock on the VCR which will run slow or fast.
Standards differ and you won't be able to watch or record broadcasts/cable
unless your equipment supports multiple standards.
The only difference between using a blank tape purchased in the US then
used in the UK is the playing time will be different ie, a T120 (2 hours)
from the US will have a longer playing/record time in the UK. This is due
to the different head drum speed i.e., 60 Hz (1800 RPM) and 50 Hz (1500 RPM).
A VCR is not simply 'analog playback' in the same way that an audio recorder
doesn't care whether you record classical or rock. The VCR must synchronize
to the video timing and demodulate the luminance and chrominance information
in order to lay down the tracks on the videotape. There are enough
differences among world video formats that while technically possible (and
such multiformat VCRs exist) it is not automatic - or free. The video
timing and modulation techniques for video formats like NTSC, PAL, SECAM,
etc. are sufficiently different that additional circuitry is necessary to
handle multiple formats. In the U.S. at least, there is not enough demand
to justify the added expense.
The technology of video recording makes interesting reading and the
sophistication of the circuitry and mechanism of a $200 VCR is quite amazing.
TVs are more likely to accommodate difference standards than VCRs.
Even a regular TV may be able to be used to play from a different
standards VCR. For example, NTSC 30/525 and PAL 25/625 use very similar
horizontal frequencies but different vertical rates and color encoding.
Playback will be possible (in B/W at least) if the vertical lock circuitry
(or the vertical hold control if there is one) on the TV has enough range.
A simple color code converter can then be easily constructed using a
couple of chips and some discrete parts.
The use of a $2 T120 tape with a HiFi VCR permits the recording of up
to 2 hours of audio with near-CD quality.
However, some designs require a video input to stabilize the drum speed
and possible degradation (e.g., wow and flutter, noise, etc.) of the
recorded audio. Some VCRs will work fine without any video. Others
need it to stabilize the drum speed from the vertical sync. For best
results of audio-only recording, find a source of video-black such as a
camcorder with the lens cap on to minimize possible video interference
(though this is usually not a problem).
(From: Mike Appenzeller (Michael.W.Appenzeller@lmco.com)).
I don't think any stereo modulators exist, other than very expensive
professional models. The processing for TV stereo sound is much more complex
than FM stereo, involving dbx companding/NR on the L-R difference signal.
Hi-Fi VCRs mix the two audio channels together before feeding a mono-audio
modulator.
I laugh at all the people who buy a Stereo TV, HiFi Stereo VCR, then insist on
using the Channel 3/4 VCR RF outputs instead of the direct A/V connections.
They are getting MONO Sound!
"Is it possible to rerecord the video (and linear audio) tracks but preserve
the HiFi audio?"
You cannot do this without disabling the erase head(s). If this is done,
you will get interference from the previously recorded video - the rainbow
patterns present at the beginning of recordings over old material on VCRs
without flying erase heads.
Even if it were possible, I don't know how robust writing over the HiFi audio
tracks would be - you might get degradation after 1 or 2 dubs.
Some VCRs do not have much of a long term memory should there be a power
failure. Can a UPS designed for a computer system be used with these VCRs
so that all the programming (and possibly channel settings as well) are
not lost every time the power line burps?
A UPS might be a solution but there are some issues to keep in mind:
* If your VCR uses a switching power supply (with no input power transformer),
it may be fine as the waveform does not matter that much. If it uses
a power transformer, then there could be problems if the waveform put out
by the UPS is far from sinusoidal - which it likely is.
* The VCR is a very light load. I don't know if this could be a problem with
some UPSs.
* The inverter in a typical UPS may generate Radio Frequency Interference
(RFI) but this probably doesn't matter if it only runs when the power fails
and you aren't viewing or recording at the time.
If your VCR recently developed this amnesia, then you might consider
attempting to locate the cause (a shorted NiCd backup battery or bad supercap)
and correcting it rather than tying up a UPS for this purpose.
The answer is yes but I would not recommend it. In order to provide
reliable backup, totally error free storage and retrieval must be
guaranteed. This is a non-trivial problem given that the video tape
is an analog storage media prone to noise and dropouts. Redundant
information would need to be stored and sophisticated error detection
and correction circuitry must be included. By the time you are done,
the theoretical capacity of a T120 video cassette of, perhaps, 5-10 GB
is greatly reduced. Furthermore, you probably want somewhat rapid
random access and this **will** be very hard on a consumer grade tape
transport designed for movie viewing and time shifting of soaps.
With the cost of reliable tape and disk storage units having enough
capacity to backup a 1 GB hard drive available or on the way for less
than $200, it doesn't make sense to use a VCR with a totally incompatible
format and questionable reliability when you will need it most - in a
data emergency.
I have no idea if the following is any good - I kind of doubt it - but
various products of this type were developed before devices like cartridge
tape (and now the Zip(tm) drive) backup became popular.
(From: Robin Gilham (gilham@stb.dfs.co.za)).
I saw an ad for a plug in card and software only yesterday, claiming
2G of storage on a 240 minute VHS tape. Wait.... yes, here it is.
The product is called "BACKER", and in .nl available from Timtronics
(+31-50-314 0937). Comes with interface card and Windoze software, will
backup harddisks at speeds of up to 9MB per minute for DFL159
(which is less than US$100).
(From: then@superpallo.cs.hut.fi (Tomi Holger Engdahl)).
The manufacturer of BACKER is Danmere Technologies Ltd and they have
WWW-pages at http://www.danmere.com/. What they claim by 9 MB per
minute, that is the data rate at the maximum speed, maximum compression
and minimum error correction. The uncompressed data rate is 5 MB
per minute at highest speed.
(From: Karl-Henrik Ryden (kalle@pobox.com)).
I have one of their cards. It works, but is rather tedious to use. It is
kind of like in the old ZXSpectrum/VIC64 days. :-)
First, you might be able to repair the VCR and prolong its life. Why are
you reading this section and considering such a transgression? Grrrr.
OK, so you really want to just use its timer. There are two things you
would have to do:
1. Trick the transport into thinking there is a recordable tape in place.
This is not hard - an old cassette shell will probably be all you need
for this. Then you don't have to worry about your non-recorded tape
from running out or wearing out.
2. Find a signal that can be used to control a relay, solid state relay,
or optoisolated triac. If you are electronically inclined, this should
not be too hard. If nothing else, the record LED or any switched power
bus would suffice. A solid state relay or optoisolated triac takes a
logic signal and will control a resistive AC load. Check the specifications
if you want to control some other type of load like a motor or external
tape deck. A suitably rated normal relay could also be used but a driver
circuit may be needed to power the coil.
Some (rare) VCRs have a switched outlet in which case this is trivial.
Perhaps, you have this fantasy:
"I'm wondering if it's possible to take the tuning circuit (tuner and
associated circuitry) out of a VCR and somehow controlling it with a PC (say
through the parallel port), and then feeding the composite signal to the
input of a video capture card?"
If removed from the VCR, you will need the complete specs on the digital
interface between the VCR's system controller and the tuner (assuming it isn't
on of those old types selected by mechanical switches!), as well a substitute
power supply. This information may not be available even if you purchase the
complete service manual. However, you may be able to infer it by monitoring
the relevant signals with an oscilloscope or logic analyzer :-).
An easier approach may be to use the entire VCR intact and interface via
the front panel (by simulating the Chan +/-, TV/VCR, etc. buttons) or via IR
by simulating its remote control.
The question you originally asked might have been: What is this alien
looking thing in my VCR?
The object in question may look like a pentagonal shaped frosted glass
slab with two pairs of wires sticking out of adjacent edges. What it is,
is an acoustic delay line implementing a one TV line (1H) delay - around 63
microseconds (NTSC). The crystal is a shaped cavity and the polished edges
are acoustic reflectors. There are a pair of piezoelectric transducers -
one to launch a wave and the other to pick it up. The acoustic waves bounce
around in a zig-zag pattern which increases the effective path length, thus
the unusual shape.
Uses in a VCR include a comb filter and tape dropout masking.
The comb filter is part of the chroma circuitry and computes the sum of
the current and previous video lines during recording and playback.
The acoustic delay line therefore implements a delay of exactly one
horizontal line. Due to the various games that are played with chroma
signal phase in the NTSC-VHS system (as well as BETA and PAL), the end
result is that chroma signal amplitude is doubled and crosstalk between
adjacent tracks is canceled out. This is because the chroma signal is
always exactly in phase between successive video lines but the crosstalk
between adjacent tracks is always exactly out of phase. The name 'comb
filter' is derived from the shape of the frequency response of the comb
filter - its evenly spaced spikes look somewhat like a hair comb and
it is used to 'come out' the crosstalk components of the chroma signal.
Another use for am acoustic delay line is dropout masking. The surface of
the tape is not always perfect - bits of oxide fall off or slight dips or
bumps result in momentary loss of head-tape contact. One way to minimize
visible streaks in the video is to replace the lost signal with video
from the previous scan line.
Nothing alien about it, just not your everyday electronic part.
(From: Jeroen H. Stessen (Jeroen.Stessen@ehv.ce.philips.com)).
There is no need for a comb filter in a camcorder!
Signals from the CCD are not first combined into CVBS to then be
separated by a comb filter again. That would make no sense.
(However, since modern camcorders are full function VCRs without a tuner, this
function may still be needed for dealing with external video input. --- sam)
Other applications for delay lines are drop-out compensation and delay
equalization between luminance and chrominance.
Did you ever wonder what happened to the ultrasonic glass delay lines that
were once used by the millions in every PAL television, for U/V separation?
They were replaced in nearly *all* applications by the Philips switched
capacitor delay lines TDA4660(-61,-62).
From: "David Lawson"
the color signal(chroma) takes longer to process than does the b/w or
luminance channel so to get the color to line up with the b/w picture, the
B/W signal has to be delayed slightly.
You have probably been unable to sleep at times thinking about this subject!
They are similar - perhaps identical in some cases as the terms both
mean the same thing. If there is no optical output/input, then they
are likely the same type of device.
Optoisolators are used the switching power supplies to couple the feedback
from the low voltage to the line-connected (non-isolated) input.
With 4 leads, these are a combination of a an LED and photodiode or
phototransistor.
With 6 leads, there may be additional circuitry providing a logic level
output, or base connection to the phototransistor, or just extra unused pins.
Photo interrupters or reflective sensors are used for detecting reel rotation
cassette presence, and mechanism position. In this case the optical path -
either direct or reflective - is external to the device.
A datasheet will clarify any functional or circuit details.
Photo interrupters or reflective sensors are used for detecting reel rotation
Testing is accomplished (1) with a multimeter for shorts or opens on the
LED and (2) by providing drive to the LED and checking the resistance of
the photodiode or phototransistor with a multimeter - it should go down
dramatically if the LED is on.
Also see the sections on sensors and sensor testing.
Don't expect an amazing answer - this is a set of questions.
Why are there so many totally different designs to do basically the
same thing? I fully understand the pressures of cost and manufacturability.
However, it would seem that with VCRs, for example, every manufacturer (of
which there are only a limited number who actually manufacture the
tape decks themselves) and every couple model years has a totally unique
design. You would think that after almost 20 years of manufacturing VHS
decks, the technology would be mature. True, there have been advances
with respect to quick start, HiFi, and so forth. Nonetheless, the required
functions have not changed. And, for that matter, the performance of the
typical mechanical deck has not improved that much in the last 10 years or
so. If anything, the old designs seem to be remarkably robust. I can
keep a 10 year old machine going virtually forever by replacing the rubber
every few years. I am not sure that I can say the same of a modern VCR.
Is it only a matter of maximizing performance at a given cost or is there
something more? NIH syndrome? Maintaining control over repair parts
and service? Or, use of entry level engineers who might provide a new
outlook on the design?
After taking your totally dead VCR into an authorized service center, it is
a month and still no diagnosis. When pressed, they finally 'discover'
that a diagnosis has been made and the estimate is $80.
The repair place is jerking you around. It should not take them as long as
you have experienced to make a diagnosis - especially if they are authorized
and have the service manual. They like the really easy problems like
"My VCR started eating tapes last week. Is it hopeless?" 50 cents worth of
rubber (idler tire), charge $50 - easy money. And they appear to be heros.
To fix the electronic problems you need at least the intelligence of a carrot
and time - and time is money. OK, so maybe they give a quick cleaning also.
If it were my VCR, I would bitch, moan, claim poverty, threaten to report
them, etc. But, get it back and fix it myself. I assume you checked
the fuses. $80 dollars to fix doesn't sound like it could have been
much more than a fuse. With the typical markup on parts (4:1 for small
parts), those alone could easily push the bill to more than $80. The
longer they hold it, the tougher the problem seems so that when presented
with the (larger) bill the customer figures it is justified.
The following is a true story. It appears to be an example of incompetence
compounded by a lack of basic decency in dealing with the customer.
> Recently my 4-5 year old JVC HR-D910U (Hi-Fi Stereo) VCR stopped
> loading tapes properly. More specifically, a rubber roller which is
> lifted up and out of the way when the tape is ejected would come down
> right on top of the tape after the tape was loaded. This occurred
> because some metal guide, which moves as part of the loading sequence,
> wasn't properly pulling the tape out of the way of the (downward
> moving) roller. Other than this problem, the VCR performed normally:
> i.e., if one manually moved the metal guide to pull the tape out of the
> way and then hit "PLAY", the machine would behave completely normally
> in all modes until the tape was ejected and another tape was loaded in.
If this were a Sony, I would say that it needed a single drop of oil on
the half-loading arm shaft - which causes quite similar symptoms. Possibly
the JVC transport is similar.
At this point, there is not much wrong with the VCR - maybe a mechanical
problem like a stripped gear or the aforementioned gummed up lubrication.
It could conceivably be electrical like a dirty or worn mode switch.
However, I would go with something mechanical - and simple to identify
and repair.
> I took the machine to a local repair shop that seemed reputable (has
> been in business for a long time, does the actual repairs for local
> stores of a large consumer electronic chain, etc...).
Of course the latter is not a testimonial. Electronics chains make their
money from selling new VCRs not from repairing old ones. Therefore, they
may have incentives to discourage people from repairing their equipment
(though mucking it up is not the usual approach - simply declare it not
worth fixing - which is I guess what they did in the end).
> After charging me a $30 estimation fee (to be used towards the repair
> if I so chose), they concluded that there was something wrong with some
> gear in the loading mechanism as well as the mode switch. The price for
> the estimated repair seemed reasonable, and so I authorized them to go
> ahead. To make a long story short, after about 2 months (!) of waiting
> (they claimed to have had trouble getting the parts) they reported that
> they had replaced the parts, but the VCR still did not work. In fact,
> it now loaded properly, but didn't play well, and in general was
> confused about what mode it was in. For example, after ejecting a
> tape, the spindles that insert into the VHS tape cartridge would
> continue to spin around (as if there were a tape in there in PLAY
> mode).
They should have been able to clean the mode switch as a temporary fix
and confirmation of the problem. A broken gear would be obvious - they
should still be able to produce it for you - not that this would mean very
much as there is no way of demonstrating that it originated in your VCR.
Two months is way too long to wait for common service parts.
At this point, the timing is probably messed up - the novice bozo who was
assigned to your VCR had not read this document and violated Rule #1: always
mark all positions of mechanical components or adjustments before replacing,
removing, moving, or changing anything.
> Their claim was that now there was something wrong with the micro-
> controller on the VCR and that it was putting out some sort of
> incorrect voltages. Moreover, this problem was allegedly masked by the
> earlier problems, and only became apparent after they had performed the
> repairs they had done.
If the microcontroller is messed up, it very likely a result of what they
did. Their 'repairs' should not have made the situation worse. It used
to be possible to play a tape by helping the loading mechanism to complete
its cycle.
> In their estimation, the price of replacing the controller wasn't worth
> it, and so they wanted to just give me the VCR back, with the repairs
> that had already done (but keeping the $30 estimation fee).
A reputable place would give you a total refund, no questions asked. Even
if it was your VCR that was hopelessly screwed up from the beginning, it
was their responsibility to recognize this.
> The repair place speculates that some voltage spike must have injured
> the controller which may have coincidentally resulted in the loading
> problem. Or, another theory they proposed was that the loading problem
> caused some motor to over-strain itself in some way which caused an
> electrical problem which injured the microcontroller.
Balderdash. The original symptoms simply do not support this in any way,
shape, or form.
> My theory is that, since the VCR was normal other than the loading problem
> described, they must have screwed the machine up during the repair, but
> do not want to take responsibility for that fact, and after putting in
> a couple dollars worth of parts are happy to keep the $30 "estimation fee"
> themselves.
This is much more likely. However, there still may be nothing seriously
wrong - the gears may just need to be retimed. This may require s service
manual, some consultation with a genuine JVC technician, or even another
similar model VCR tape transport to compare it with.
> So, is their version of the story even remotely possible? If not, I feel
> that they destroyed a perfectly good machine with a minor problem and I'm
> wondering what, if any, recourse I might have in this sort of situation.
While anything is possible, I think, to put it bluntly, they do not have
a clue. Motors do not damage microcontrollers. There was nothing seriously
wrong when you took it to them - it should at least be possible to put it
back in that condition. Since they did not do this, whatever they did is
now the cause of more significant problems. However, it is quite possible
that even these can easily be remedied by proper timing of the gears and
mode switch - in addition, possibly, to that single drop of oil.
My recommendation would be to take it to an authorized JVC repair center
with this story printed out (not to blame the other people but to give them
something to start with). A competent technician should be able to quickly
determine what is going on. If they concur with your assessment of the
situation, then you can try to get your $30 back from the VCR repair shop
from Hell you have not already been credited.
The only differences in testing between a visible and IR LED (or IR Emitting
diode - IRED) are that:
* The voltage drop across an IR LED will be slightly lower - perhaps 1-1.5 V
instead of around 1.7-2 V for visible types.
* The light is not usually visible to most humans. Thus you need something
sensitive to IR. See the section: "IR detector circuit" or try a CCD
camcorder or those IR detector cards. Some people are supposed to be
able to detect light well into the IR. I am not one of them.
For in-circuit tests with power applied:
* If you measure 0 V across it, the LED is shorted or the power supply is bad
or disabled.
* If you measure greater than 1.5 V across it, the LED is open.
This IR Detector may be used for testing of IR remote controls, CD player
laserdiodes, and other low level near IR emitters.
Component values are not critical. Purchase photodiode sensitive to near
IR - 750-900 um or salvage from optocoupler or photosensor. Dead computer
mice, not the furry kind, usually contain IR sensitive photodiodes. For
convenience, use a 9V battery for power. Even a weak one will work fine.
Construct so that LED does not illuminate the photodiode!
The detected signal may be monitored across the transistor with an
oscilloscope.
Vcc (+9 V) >-------+---------+
| |
| \
/ / R3
\ R1 \ 500
/ 3.3K /
\ __|__
| _\_/_ LED1 Visible LED
__|__ |
IR ----> _/_\_ PD1 +--------> Scope monitor point
Sensor | |
Photodiode | B |/ C
+-------| Q1 2N3904
| |\ E
\ |
/ R2 +--------> GND
\ 27K |
/ |
| |
GND >--------+---------+
_|_
-
The general arrangement of audio, video, and control information is shown
below for a VCR with stereo audio. This view is from the front of the
transport (through the back of the tape):
Top edge of tape
------- ------------------------------------------- --------
^ Right linear stereo .35 mm
| ------------------------------------------- --------__ (Guard band,
| ------------------------------------------- -------- .3 mm)
| Left linear stereo .35 mm
| =========================================== ========
| \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ <-- Start of scan,
| \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Top of picture,
1/2" \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Left roller guide.
12.7 mm \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \
| \ \ \ \ \ Video with HiFi sound \ \ \ \ \ \ Tape motion -->
| \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \
| \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Right roller guide,
| \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ Bottom of picture,
| \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ <-- End of scan.
| =========================================== ========
v Control track .75 mm
------- ------------------------------------------- --------
Bottom edge of tape
Note: ==== denotes the .15 mm guard bands between video, and the audio and
control tracks. Thus, once the audio, control, and guard bands are taken
into consideration, only about 10.65 mm or .42 inches is available for the
diagonal video tracks.
For a VCR with HiFi audio, the HiFi audio heads travel the same path as the
video heads but record their information just before the video heads pass
over the same spot on the tape. Although some of this is then partially erased
by the video, enough remains deep in the tape oxide to to permit reconstruction
of CD quality sound. The difference in azimuth angles of the video (+/- 6
degrees) and HiFi audio heads (+/- 30 degrees) minimizes interaction.
For a VCR with monophonic audio, the left and right audio tracks and their
guard band are combined into a single audio track of about 1 mm width.
Dimensions are most definitely *not* drawn to scale. The Audio and control
tracks are very narrow in comparison to the tape width. To get an idea of the
actual slant angle of the video tracks, imagine the tape stretched horizontally
by about a factor of about 10. (The length of a video track representing one
field or 262.5 scan lines is about 3.8 inches.) There are also actually more
than a hundred tracks at any given location side-by-side across the less than
.42" available for the video information. This number of tracks is equal to
175 at SP, 350 at LP, and 525 at EP speed (for NTSC 525/60 - note that this is
not a coincidence but that is another story). Think of the alignment precision
needed for proper tracking! You can estimate this number by just timing how
long it takes for the rainbow pattern to wipe down the screen when re-recording
over an old tape at either speed on a VCR without flying erase heads and
multiplying this time by 60.
Go to [Next] segment
Go to [Previous] segment
Go to [Table 'O Contents]