Binatone TV Master MK10

[spacebiscuit]

Hi,

I recently picked up a Binatone TV Games Master MK10 as a project to experiment with for nostalgic reasons:

https://www.ithistory.org/db/hardware/binatone-telecom/binatone-tv-master-mk-10-01-4834

I read about a composite video mod that by-passes the original RF output enabling the console to be hooked up to a modern TV. This is the mod I had printed and which I put together (note the project is for an Atari but the principle works here too):

https://www.pcbway.com/project/shareproject/Thumbnail_board_...7800_RF_composite_modification.html

I've installed the Mod and I have output to my TV. However there are a few issues that I am trying to iron out, not all related to the mod.

1. The composite signal is subject to some rolling interference, sometimes it is worse than at other times. I am wondering if it is a shielding/earth issue?

2. A couple of the game selectors do not work, I've tried de-soldering the selector and it didn't appear to be defective

3. The board has 2 x 220uf 10v caps and 2 x 100uf 10 caps - any idea what their purpose on this board. I can't be certain that they haven't been switched around when this was re-capped (long story). What would be the significance of switching the 100uf and 200uf caps if they are in the wrong positions?

I've attached a copy of the wiring diagram that is included in the manual - but with my limited electronics skills I am not finding it the easiest to read!

[SiriusHardware]

Hi spacebiscuit - taking your points in no particular order:-

A couple of the game selectors do not work, I've tried de-soldering the selector and it didn't appear to be defective

Would those be the Target 1 and Target 2 games, by any chance? Both of those games would need a (usually optional) light-gun attachment to work, and incidentally they would also need a real CRT TV display to work as well.

The board has 2 x 220uf 10v caps and 2 x 100uf 10 caps - any idea what their purpose on this board. I can't be certain that they haven't been switched around when this was re-capped (long story). What would be the significance of switching the 100uf and 200uf caps if they are in the wrong positions?

Three out of four of those capacitors are directly across supply rails, their purpose is to provide local filtering / stabilisation of the supplies and if they were in the wrong way with the unit powered up for any length of time they would soon show signs of distress by bulging, leaking, heating up or even going bang. What all four have in common is that their -ve pins should all go to circuit ground / 0V, so if you have a meter, use it to check for continuity / very low resistance between ground / 0V and the -ve pins of all four of those capacitors.

Thanks for posting the circuit - we are always interested to see things like this. The modulator made from discrete components seems quite unusual, it's not the ASTEC modulator in a tin can that we would usually expect. It also looks as though the chip has multiple video outputs, probably intended to be encoded as colours, but here they, along with the sync signal, appear to be mixed together by different combinations of resistors to produce a grey-scale video output. (Not just black and white, but a number of shades of grey as well).

Where exactly on the circuit diagram did you tap off your feed to the composite buffer stage you added?

[spacebiscuit]

Thanks for the reply - the console came in the original box with original packaging and had the manual which helpfully has he diagram I posted.

I believe the MK10 was also available in a special "colour" variant which I believe was achieved with some trickery on the circuit board to provide an intrarpretation of colour a output. I think you're right about this model providing grey-scale output which I guess was an upgrade on the black/white output of earlier models but not quite the "colour" output - something in between?

I have tapped of that convergence of resistors in the center of the board (see my yellow wire). I took the ground from the top right (black), +5v from the red (also top rightish) and sound white from the top rightish also.

[Realtime]

Hi. Colour versions used a companion chip (AY-3-8615 “super star color converter”) to generate the colour signal. The AY-3-8610 provides the luminance (grey scale) and timing signals. Attached shows the pin names for the signals that form the composite video signal to the Modulator. If you are tapping off using a transistor buffer you may need to bias the transistor base to get the best signal.

This chip is covered at page 26 of the following. If you need more info let us know.

http://bitsavers.org/components/gi/_dataBooks/GIMINI_TV_Game_Circuits_197801.pdf

[SiriusHardware]

I think this is Spacebiscuit's cue to post photos of both sides of the buffer circuit PCB, or post a quick sketch of the circuit of it with the positions and values of the resistors around the transistor drawn on. A photo of a simple pencil sketch of the buffer diagram will do, as long as it is readable.

[spacebiscuit]

Here is the mod board that I made, the parts are as follows:

Q1: 2N3904
R1: 2.2k ohm
R2: 3.3k ohm
R3: 75 ohm

A question I missed earlier, the games which won't select are both on the 2nd row, they are not the target games. The target games do start/select.

Thanks.

[SiriusHardware]

So that looks like the attached, if I have 'read' it correctly.

Any thoughts as to resistor values, Realtime?

[Realtime]

Wrt resistor values, the 2K2/3K3 resistor bias is resulting in about 3V DC offset at the video buffer output. That's way too high and may cause the monitor some problems, depending on how it clamps the incoming video.
Try the changes shown in the attached pdf. A couple of things to note:

1) make sure you're taking the power for the video buffer board from the regulated side of the transistor (~6V). This will ensure the video output is biased at around 0.4V dc
2) Remove the 1K8 resistor shown as crossed out in the PDF (or lift one side of it). This will remove any load presented by the modulator and remove any RF feed through.
3) Replace R1 (2K2) with 15K to reduce the bias to about 1vdc

This is a bit of a stab in the dark, because the modulator circuit presents a non-linear load and so the ratio of video to sync amplitudes will be 'wrong' when seen by the monitor, but the fact you're getting a picture already suggests that won't be a problem.

If you find you're getting no signal try reducing R1 in steps (e.g 12K, 10K, 6K8, etc) to adjust the bias.

[ortek_service]

I'm surprised that the exact DC bias-level of the composite video signal into a (modern at least) TV is that critical, and just biasing it mid-supply, to avoid clipping would be OK if it is actually AC-coupled (then DC-restored) by the TV's SCART etc inputs.

As I read a while ago - from here: https://www.analog.com/en/technical-articles/get-a-grip-on-c...as-and-accoupled-video-signals.html
That:

Despite these complications, video signals must be AC-coupled at the point where the voltage domains change. Connecting two different power sources through a DC connection is dangerous, and usually forbidden by safety regulations. Consequently, video-equipment makers have a tacit agreement to AC-couple the input of their equipment, and DC-couple the output—requiring the next stage to re-establish the DC component (see EN 50049-1 for PAL/DVB [SCART] and SMPTE 253M section 9.5 for NTSC, which allow a DC output level). Failure to establish such a protocol leads to "double coupling," where two coupling capacitors appear in series, or to a short where there are no capacitors. The sole exception to this rule is battery-powered equipment, such as camcorders and still cameras, which AC-couple the output to minimize battery drain.

- Although maybe monitors that pre-date these standards, may not have been AC-coupled etc.

[Realtime]

Originally Posted by ortek_service

I'm surprised that the exact DC bias-level of the composite video signal into a (modern at least) TV is that critical, and just biasing it mid-supply, to avoid clipping would be OK if it is actually AC-coupled (then DC-restored) by the TV's SCART etc inputs.

I agree it's not critical and probably will work with most TVs, although I have a 5" B/W TV that makes a good bench monitor but that one is not ac coupled so there are exceptions. I usually just stick a big electrolytic in line and that works well (220uF is big enough to avoid video line droop).

Of course, it's better to reduce large dc offsets where they're not needed to reduce power dissipation and battery drain. With a 3V offset the 75R resistor is passing 40mA continuously; with a 0.4V offset it's down to 5mA. Every little helps

[spacebiscuit]

I'm try to ensure i have the 4 caps in the right positions.

Can anyone help me to identify which should be in each position.

They should be 2 x 100uf and 2 x 220uf. I've marked them on the attached photo a 1-4 and 3 of the transistors.

I've been trying with multi-meter to identify which is which but I'm going around in circles.

Thanks.

[Realtime]

Looks like

1: 220uF (Bulk decoupler for input power to 9013 collector)
2: 220uF (Connects to 9015 emitter - audio coupling)
3: 100uf (Bulk decoupler for the modulator- large metal can)
4: 100uF (Bulk decoupler for 6V supply from 9013 emitter)

Attached is the mirror view of the tracking to make it easier to match up with the components on the top side.

[spacebiscuit]

Originally Posted by Realtime

1) make sure you're taking the power for the video buffer board from the regulated side of the transistor (~6V). This will ensure the video output is biased at around 0.4V dc

I'm just trying this - is the regulated side of the transistor as shown in my attached photo (where i have soldered already the red wire)? The photo orientation is as you'd be looking at it per the other photos. Can I confirm by checking I have 6v there?

Thanks.

[ortek_service]

Originally Posted by spacebiscuit

I'm just trying this - is the regulated side of the transistor as shown in my attached photo (where i have soldered already the red wire)? The photo orientation is as you'd be looking at it per the other photos. Can I confirm by checking I have 6v there?

Thanks.

Yes, that does appear to be correct to me - as looking at some datasheets for the 9013(G) transistor, the emitter is the left-most pin when looking at with markings facing you and legs facing downwards.
And the track from this should also go to pin 14 of the IC / positive of a 100uF Electrolytic capacitor etc.

Plus, You should also measure approx 6.0V there, as long as there's sufficient external supply voltage input for the regulator to function OK.

[Realtime]

Originally Posted by ortek_service

And the track from this should also go to pin 14 of the IC / positive of a 100uF Electrolytic capacitor etc.

That should be pin 16, but you're right, you should measure about 6V on that pin.

[ortek_service]

Originally Posted by Realtime

That should be pin 16, but you're right, you should measure about 6V on that pin.

Yes, I'd seen it went to one-up from pin 15 at the bottom left corner of the IC symbol - so I'd expected that to usually be pin14.

But I've just noticed although they've used a DIL package for the IC symbol, they've actually (rather-unconventionally) numbered it as an underside-view!
So they should probably have numbered all the pins, rather than just the corner ones, to make it absolutely-clear they weren't swapping any others around.

[SiriusHardware]

Following up on another issue with the unit, which two games weren't starting when the relevant buttons were pressed? (I know spacebiscuit already said it wasn't the light gun games, but I can't recall any mention of which ones weren't working).

[spacebiscuit]

I took the power from the 6v pin of the transistor as advised - which I confirmed with a multi-meter. although there was less interference and the rolling bars are gone there is less greyscale - the output is pretty much blank and white resulting in one of the player's bat/paddle is now difficult to distinguish from the background and almost appears to be invisible.

If I switch to the 12v side of the transistor (presumably the input) the greys appear as I would imagine they would have originally, player 1 is white, player 2 black and the court/background is grey - but there is the significant interference as described previously. See the attached photo that shows the rolling bars.

I didn't yet try lifting the leg of the resistor - is that significant if I am using the 6v and the video I am still taking from the confluence of solder joints per my photo.

[SiriusHardware]

We might be better able to advise you if we could 'see' the video output waveform that is being produced by your buffer circuit. The overall peak-to-peak voltage of the video waveform is supposed to fall within a fairly well defined range and the relative sizes and levels of the sync portion of the waveform, the black level and the peak while level are all supposed to have a particular proportional relationship to each other.

The buffer circuit you have there was probably specifically designed for the Atari console and the resistor values chosen to produce a video signal of appropriate dimensions when connected to that type of unit, so for the buffer circuit, unaltered, to produce roughly the right size and shaped waveform when connected to something completely different is really just a matter of luck. Might work, might not.

All of which leads to the question: Do you have, or do you have access to, an oscilloscope?

[ortek_service]

Originally Posted by spacebiscuit

I took the power from the 6v pin of the transistor as advised - which I confirmed with a multi-meter. although there was less interference and the rolling bars are gone
>>
>>
If I switch to the 12v side of the transistor (presumably the input) the greys appear as I would imagine they would have originally, player 1 is white, player 2 black and the court/background is grey - but there is the significant interference as described previously. See the attached photo that shows the rolling bars.

The rolling-bars are most-likely 'hum-bars' caused by 50Hz / 100Hz ripple on the main (unregulated?) external supply input
- Not sure if you're using the original mains-PSU 'AC-Adapter', or a substitute, but it is probably an unregulated type (may also be not very well-smoothed, if internal regulated normally deals with the ripple as long as voltage doesn't dip below approx 7V).
So using a better PSU could solve that.

However, it really shouldn't be necessary to run the buffer-amp off 12V - And may be exceeding ratings of some of the components if it is run that high, without changing resistor values accordingly.

So ensuring that the buffer-amp is correctly-biased, with values previously suggested, should enable it to be run off 6V regulated supply.

It may also be worth putting a few 10's of uF capacitor in-series with the output (Probably positive of an electrolytic etc. to the output of the buffer circuit), just to see if your TV/Monitor is getting upset by DC-level etc at the output.
There may also be a level issue, if the greyscale isn't right, but really need to check this with a 'scope (even a very-cheap low-bandwidth one, should be able to see the video signalsl, to be sure about this (plus check if TV/Monitor is terminating the output with 75R etc, so will be reducing level by half).