Replacment for IC401

[Cecil Casey]

For some reason this did not post the first time I sent it. So here goes
again...

I am looking for anyone that has a dead vectrex that has the typical IC401
(the LM379) problem. As you all know the LM379 has been out of production
since 1983 or so, but I believe that I have located a suitable replacement
part. But first off, a little background on the circuit in question.

IC401 is the current source op/amp for the X and Y axis deflection coils.
Each of the axis use 1/2 of the LM379. The LM379 has been known to be prone
to failure for years. This is documented as a common hardware failure in
both Vectrex, as well as upright arcade machine sound boards. Failure
symptoms of the LM379 on the Vectrex are as follows.

The machine will start with no display unless the brightness knob is turned
up to max, when you will see a single dot in the center of the display. All
sounds will be reproduced normally. The reason you only see the dot at the
highest intensity settings is that the beam blanker kicks in when there is
no X-axis (horizontal) deflection.

You can also get a slightly different display if only the Y-axis side of the
LM379 has failed. In this case you still get normal sound. The display
will show a single raster line centered vertically on the screen. In other
words you will only see a single line reaching from the right to left side
of the screen. Typically this line will have 'dancing dots' on it as
objects are being drawn to the screen. You get this display instead of the
blank picture as before due to the deflection detection circuits that enable
J-FET's 401 and 402 are only taking their sense signal from the horizontal
deflection.

I am looking for anyone that has a vectrex with the above failures that
feels comfortable with a soldering iron, and does not mind building simple
circuits. The reason this is necessary is that there are no 'pin for pin'
replacements available for the LM379. So to repair requires you to remove
the old chip and replace it with a small daughter board that holds the new
IC and runs short wire leads to the old chip's socket.

So now that I have gone and written what everyone here most likely knows
already :)

I believe that National's newer part, the LM4765 will make a good
replacement for the LM379. The problem is not at the signal levels, it is
physical. The LM379 is a 14 pin DIP, but the LM4765 is a 15 pin TO-220
package. This will require placing the LM4765 on an adaptor board. The
respective pinouts from the data sheets are shown below.


|--------U---------|
Vcc 1 -| |- 14 (Vcc-Vee)/2
| |
| |
NC 2 -| |- 13 NC
| |
| L |
SIG GND 3 -| M |- 12 SIG GND
| 3 |
| 7 |
PWR GND 4 -| 9 |- 11 POW GND
| |
| |
OUTPUT 1 5 -| |- 10 OUTPUT 2
| |
| |
INPUT 1 (inv) 6 -| |- 9 INPUT 2 (inverting)
| |
| |
FEEDBACK 1 NI 7 -| |- 8 FEEDBACK 2 (non
inverting)
| -----------------|



Now compare that to the pinout on the LM4765...

/--------------------
| |
| |-- 15 Vcc SIDE B
| |
| |-- 14 STANDBY B
| |
| |-- 13 FEEDBACK B (NON INVERTING)
| |
| |-- 12 INPUT B (INVERTING)
| |
| L |-- 11 MUTE B
| M |
| 4 |-- 10 GND B
| 7 |
| 6 |-- 9 STANDBY A
| 5 |
| |-- 8 FEEDBACK A (NON INVERTING)
| O |
| |-- 7 INPUT A (INVERTING)
| |
| |-- 6 MUTE A
| |
| |-- 5 GND A
| |
| |-- 4 Vee
| |
| |-- 3 OUTPUT A
| |
| |-- 2 Vcc SIDE A
| |
| |-- 1 OUTPUT B
\-------------------|

As you can see there are a few extra pins on the LM4765, but it accepts the
same voltages as the LM379, it has nearly identical open loop gains(not that
it matters as we are using it in a current feedback mode anyhow), as well
as identical slew rates for the amplifiers.

As good as this seems, you also get better signal to noise ratio, better
linearity across the usable range, as well as better channel separation than
in the LM379. And the best part is that it is still in production! I have
a little converter circuit that I would like to try, so if there is a
Vectrex out there that need a fix drop a note here on the NG.

Hope you find this useful.

--
=-=-=-=-=-=-=-=-=-=
-Cecil Casey

[John Dondzila]

Post by Cecil Casey
I am looking for anyone that has a vectrex with the above failures
that feels comfortable with a soldering iron, and does not mind
building simple circuits. The reason this is necessary is that there
are no 'pin for pin' replacements available for the LM379. So to
repair requires you to remove the old chip and replace it with a small
daughter board that holds the new IC and runs short wire leads to the
old chip's socket.

A few years ago I honked off FRW by finding a suitable replacement
for the LM379 when he was trying to sell a beat up pull on Ebay
for some outrageous price (Still out there buddy ? Have you bought
a plot alongside your own to bury your collection ?).

I haven't been able to find the post on Deja, but I came across a site
which made replacements for old chips. I think they were actually PALs
or PICs which "simulated" the chip.

I'll look around some more and post it if I find the old link/site.

JD

[John Dondzila]

Is this possible to adapt to a Vectrex ? I didn't
really read into it.

http://www.geocities.com/kirbseepe/repairAMP.html

Bob Roberts has LM379S, $10

http://www.dameon.net/BBBB/parts.html

[Cecil Casey]

Post by John Dondzila
Is this possible to adapt to a Vectrex ? I didn't
really read into it.
http://www.geocities.com/kirbseepe/repairAMP.html

I went and took a look at this design, and while that design was sutiable
for sound amplification it would not be so useful in the vectrex. The
origional LM379 was a 2 channel stero amp built on one chip. So you would
have to build seperate circuits for the x and y deflection channels, then
you would have to heat sink both chips seperately.

Post by John Dondzila
Fred did one using the LM1875 a while back...
http://home.mchsi.com/~jstancliff/wsb/media/26501/site1060_t.jpg
-Clay

The same problem applies to the link that Clay posted, that chip is still
only a mono chip. There is a reason you only saw mono amps for years after
the stero lm379, they just dissipated to much heat in to small an area, and
they would cook themselves. That is what excited me about the LM4765, it is
stero and uses the same voltage ranges as the origional 379. If you could
manage to bend the pins to fit into the holes in the socket for the 379 and
put a heat sink on it the Vectrex would never know the difference!

Post by John Dondzila
Bob Roberts has LM379S, $10
http://www.dameon.net/BBBB/parts.html

For now, being that the parts are still reasonable(I guess $10 plus shipping
for a 20 year old chip is good?) there is no reason to worry about this
much. I just was doing this for fun, if someone finds it useful, all the
better! Just to let you know you can get free samples of the LM4765 from
national.com, a heatsink and a piece of protoboard would be about $5. The
only downside to doing that is now your Veccy is not original stock. But I
get the feeling that as we work along and keep this baby alive that stock
will start to mean much less than working.

Post by John Dondzila
Did a quick google of surplus suppliers -
http://www.cpcares.com/LM.html
http://www.stuutenbruin.com/linea/lineair10.html
http://www.jmb-electronique.com/INDEX%20Ampli%20audio.htm
http://www.techsonicusa.com/pages/827252/
I have a couple on standby, for when my Vec's start dieing :(

That first supplier had good (under $6) prices for the LM379's. The others
I had less luck getting prices off the sites, but if you were in Germany or
France they would be worth a look, as they both did list the part.
Techsonic listed the part but when I put it in the search it failed, I
should call them and see.

Thanks for all the responses, if there is interest I can make up a quick
line drawing to try and illustrate what pins on the LM4765 would have to be
connected to the corresponding pins on the LM379 socket.

Catch ya all later...

[juice28]

Cecil,

Your idea for the LM4765 will probably work, but I bet you will find
that it won't work as a drop in replacement. For some reason these newer amp
chips get really pissed when you run the inverting or non-inverting input
right to ground. I ended up having to use the cap to ground for it to work
with the lm1875T circuit I built. Anyone have any ideas on this? I never
investigated the internals enough to see why this was happening, but there
is definetly a difference between the 379 and these newer amps.

Fred

Post by Cecil Casey

Post by John Dondzila
Is this possible to adapt to a Vectrex ? I didn't
really read into it.
http://www.geocities.com/kirbseepe/repairAMP.html

I went and took a look at this design, and while that design was sutiable
for sound amplification it would not be so useful in the vectrex. The
origional LM379 was a 2 channel stero amp built on one chip. So you would
have to build seperate circuits for the x and y deflection channels, then
you would have to heat sink both chips seperately.

Post by John Dondzila
Fred did one using the LM1875 a while back...
http://home.mchsi.com/~jstancliff/wsb/media/26501/site1060_t.jpg
-Clay

The same problem applies to the link that Clay posted, that chip is still
only a mono chip. There is a reason you only saw mono amps for years after
the stero lm379, they just dissipated to much heat in to small an area, and
they would cook themselves. That is what excited me about the LM4765, it is
stero and uses the same voltage ranges as the origional 379. If you could
manage to bend the pins to fit into the holes in the socket for the 379 and
put a heat sink on it the Vectrex would never know the difference!

Post by John Dondzila
Bob Roberts has LM379S, $10
http://www.dameon.net/BBBB/parts.html

For now, being that the parts are still reasonable(I guess $10 plus shipping
for a 20 year old chip is good?) there is no reason to worry about this
much. I just was doing this for fun, if someone finds it useful, all the
better! Just to let you know you can get free samples of the LM4765 from
national.com, a heatsink and a piece of protoboard would be about $5. The
only downside to doing that is now your Veccy is not original stock. But I
get the feeling that as we work along and keep this baby alive that stock
will start to mean much less than working.

Post by John Dondzila
Did a quick google of surplus suppliers -
http://www.cpcares.com/LM.html
http://www.stuutenbruin.com/linea/lineair10.html
http://www.jmb-electronique.com/INDEX%20Ampli%20audio.htm
http://www.techsonicusa.com/pages/827252/
I have a couple on standby, for when my Vec's start dieing :(

That first supplier had good (under $6) prices for the LM379's. The others
I had less luck getting prices off the sites, but if you were in Germany or
France they would be worth a look, as they both did list the part.
Techsonic listed the part but when I put it in the search it failed, I
should call them and see.
Thanks for all the responses, if there is interest I can make up a quick
line drawing to try and illustrate what pins on the LM4765 would have to be
connected to the corresponding pins on the LM379 socket.
Catch ya all later...

[Zonn]

Post by juice28
Cecil,
Your idea for the LM4765 will probably work, but I bet you will find
that it won't work as a drop in replacement. For some reason these newer amp
chips get really pissed when you run the inverting or non-inverting input
right to ground. I ended up having to use the cap to ground for it to work
with the lm1875T circuit I built. Anyone have any ideas on this?

A quick glance shows that your running in a non-inverting mode whereas the
Vectrex used inverting amplifiers. Using your circuit as a direct replacement
would cause the images to be upside down and backwards, unless you also reverse
the yoke connections.

You've also removed the linearity circuits which will cause the images to bloom
somewhat as they approach the edge of the screen.

Also missing is part of the spotkiller circuit which can lead to a small burn
mark in the center of the screen after too many power cycles.

Post by juice28
I never
investigated the internals enough to see why this was happening, but there
is definetly a difference between the 379 and these newer amps.

You are missing the output stabilizing "resister/capacitor to gnd" circuit shown
on the "Typical Example" which is usually there to keep highly reactive loads
from causing things to oscillate. They are present on the original Vectrex
schematic and are missing on yours.

The Vectrex also does not connect it's inputs directly to ground, but uses a
3.3k resistor, bypassed by a .1uf cap, which maintains the DC response of the
amplifier. Your series capacitor removes this DC connection.

Both amplifiers are "self centering" and that may be part of the problem when
connecting an input directly to ground. (This is just a guess.)

I have no idea if any of the above will effect the stability of a direct gnd
connection. Though it seems like without a direct gnd connection you are not
running a DC amplifier, which (if this is truly the case), will cause image
drift. If more vectors are drawn on one side of a screen than the other, the
side with the larger amount of vectors will drift towards the center as the
unmarked capacitor next to the 1k resistor attached to pin 2 charges. But this
may be invalidated by the "self centering" circuitry, since I don't know how
that works.

-Zonn
--------------------------------------------------------
Zonn Moore
Zektor, LLC
www.zektor.com

Remove the ".AOL" from the email address to reply.

[juice28]

Post by Zonn
You are missing the output stabilizing "resister/capacitor to gnd" circuit shown
on the "Typical Example" which is usually there to keep highly reactive loads
from causing things to oscillate. They are present on the original Vectrex
schematic and are missing on yours.

Tried different combinations of this and it made no effect on the signal so
I just left it off.

Post by Zonn
A quick glance shows that your running in a non-inverting mode whereas the
Vectrex used inverting amplifiers. Using your circuit as a direct replacement
would cause the images to be upside down and backwards, unless you also reverse
the yoke connections.

Right. When I tried running it in inverted mode it had major problems.
Anytime you ran the input directly to ground (even through a resistor) the
amp got hotter then crap. I just reversed the yoke connections on my test
circuit.

Post by Zonn
You've also removed the linearity circuits which will cause the images to bloom
somewhat as they approach the edge of the screen.
Also missing is part of the spotkiller circuit which can lead to a small burn
mark in the center of the screen after too many power cycles

Just a Proto Geez ;)

Post by Zonn
I have no idea if any of the above will effect the stability of a direct gnd
connection. Though it seems like without a direct gnd connection you are not
running a DC amplifier, which (if this is truly the case), will cause image
drift. If more vectors are drawn on one side of a screen than the other, the
side with the larger amount of vectors will drift towards the center as the
unmarked capacitor next to the 1k resistor attached to pin 2 charges. But this
may be invalidated by the "self centering" circuitry, since I don't know how
that works.

It never drifted, but I thought the same thing hence trying to get one or
the other inputs to ground.

Fred

Post by Zonn
On Thu, 16 Oct 2003 21:51:27 GMT, in msg

Post by juice28
Cecil,
Your idea for the LM4765 will probably work, but I bet you will find
that it won't work as a drop in replacement. For some reason these newer amp
chips get really pissed when you run the inverting or non-inverting input
right to ground. I ended up having to use the cap to ground for it to work
with the lm1875T circuit I built. Anyone have any ideas on this?

A quick glance shows that your running in a non-inverting mode whereas the
Vectrex used inverting amplifiers. Using your circuit as a direct replacement
would cause the images to be upside down and backwards, unless you also reverse
the yoke connections.
You've also removed the linearity circuits which will cause the images to bloom
somewhat as they approach the edge of the screen.
Also missing is part of the spotkiller circuit which can lead to a small burn
mark in the center of the screen after too many power cycles.

Post by juice28
I never
investigated the internals enough to see why this was happening, but there
is definetly a difference between the 379 and these newer amps.

You are missing the output stabilizing "resister/capacitor to gnd" circuit shown
on the "Typical Example" which is usually there to keep highly reactive loads
from causing things to oscillate. They are present on the original Vectrex
schematic and are missing on yours.
The Vectrex also does not connect it's inputs directly to ground, but uses a
3.3k resistor, bypassed by a .1uf cap, which maintains the DC response of the
amplifier. Your series capacitor removes this DC connection.
Both amplifiers are "self centering" and that may be part of the problem when
connecting an input directly to ground. (This is just a guess.)
I have no idea if any of the above will effect the stability of a direct gnd
connection. Though it seems like without a direct gnd connection you are not
running a DC amplifier, which (if this is truly the case), will cause image
drift. If more vectors are drawn on one side of a screen than the other, the
side with the larger amount of vectors will drift towards the center as the
unmarked capacitor next to the 1k resistor attached to pin 2 charges. But this
may be invalidated by the "self centering" circuitry, since I don't know how
that works.
-Zonn
--------------------------------------------------------
Zonn Moore
Zektor, LLC
www.zektor.com
Remove the ".AOL" from the email address to reply.

[Zonn]

Post by juice28

Post by Zonn
You've also removed the linearity circuits which will cause the images to

bloom

Post by Zonn
somewhat as they approach the edge of the screen.
Also missing is part of the spotkiller circuit which can lead to a small

burn

Post by Zonn
mark in the center of the screen after too many power cycles

Just a Proto Geez ;)

Sorry I didn't mean to come out sounding harsh! ;-)

You said: "Anyone have any ideas..." and being bored I just started comparing to
the two circuits, don't take it wrong, I can be a bit obsessive at times! :-)

It's not like I'm complaining about the work you're doing. Hey, the more work
you put into getting new X/Y monitors built, the more audience I have for our
ZVGs!

Just ignore me and keep up the good work! (Read this as: "Hurry up and get some
yokes made!". ;-)

As a side note: Now that you can legally license the ROMs from Atari, and Oscar
has re-manufactored the Tempest spinner, and with us manufacturing the PC based
Vector generator, if you come up with a new X/Y monitor, a Tempest game could be
built from scratch using all new components! And no legal hassles! (Well maybe,
if you tried to put it on location, I'm not sure how the StarRoms license
handles that...)

Now that's cool! (Tempest being my favorite arcade game!)

Ok, back to your normally scheduled thread...

Post by juice28

Post by Zonn
I have no idea if any of the above will effect the stability of a direct

gnd

Post by Zonn
connection. Though it seems like without a direct gnd connection you are

not

Post by Zonn
running a DC amplifier, which (if this is truly the case), will cause

image

Post by Zonn
drift. If more vectors are drawn on one side of a screen than the other,

the

Post by Zonn
side with the larger amount of vectors will drift towards the center as

the

Post by Zonn
unmarked capacitor next to the 1k resistor attached to pin 2 charges. But

this

Post by Zonn
may be invalidated by the "self centering" circuitry, since I don't know

how

Post by Zonn
that works.

It never drifted, but I thought the same thing hence trying to get one or
the other inputs to ground.

Probably that magic "Self Centering" stuff! Maybe it's also trying to fight
your connection to ground. If the self centering was not exactly 0v (or your
supplies were not exactly symmetrical), it could be the IC sees this offset and
starts biasing itself to correct for it, and since it's shorted to ground all it
can do is fight the short, causing the heat. The LM379 might not be as
aggressive in it's self centering. But this is nothing but a *wild* guess!

It could also be going into oscillation, for some reason, causing it to really
heat up as well, you'd need a scope to figure out what exactly was going on.

-Zonn
--------------------------------------------------------
Zonn Moore
Zektor, LLC
www.zektor.com

Remove the ".AOL" from the email address to reply.

[juice28]

Zonn,

I didn't take it as you being harsh... Its all good.

I was just Lazy when I built this circuit. I built it more or less in a
night just to see if it was possible then I quickly lost interest (such is
the case on many things) when it worked and the theory was proved to myself
I moved on to other things.

Maybe sometime I'll revisit it with a dual amp (like the LM4765), scope
and actually do the math to see whats up haha.

It might just be better to use a couple of tip31C's and tip32C's in a
circuit similair to the Atari's for a whole new deflection circuit. Do the
differential amp deal with feedback and all. It would probably have a better
slew rate as you could jack the voltage up using the tips. Of coarse it
wouldn't just be plug and play.

I've rewound wyse yoke for use on the Vectrex and they work perfect, but
again alas I am a lazy bastard and don't want to go into the manual yoke
winding bussiness :-)

Also I wanted at some point to make a little 9" or 13" color monitor for
the Vectrex and see if someone could map the other 2 colors somewhere in
memory and build the dac circuit for them. Then some 6809 software guy's
could make color games :)

Someone make a 2 dac circuit and map it into someplace out of the way of
anything else so its compatiable and let me know. I'll put together a 9" or
13" color vector for it.

Also I gotta get one of those Zectors ;) wanna trade something for one?
New HV or deflection or something.

Fred

On Fri, 17 Oct 2003 00:15:46 GMT, in msg

Post by juice28

Post by Zonn
You've also removed the linearity circuits which will cause the images to

bloom

Post by Zonn
somewhat as they approach the edge of the screen.
Also missing is part of the spotkiller circuit which can lead to a small

burn

Post by Zonn
mark in the center of the screen after too many power cycles

Just a Proto Geez ;)

Sorry I didn't mean to come out sounding harsh! ;-)
You said: "Anyone have any ideas..." and being bored I just started comparing to
the two circuits, don't take it wrong, I can be a bit obsessive at times! :-)
It's not like I'm complaining about the work you're doing. Hey, the more work
you put into getting new X/Y monitors built, the more audience I have for our
ZVGs!
Just ignore me and keep up the good work! (Read this as: "Hurry up and get some
yokes made!". ;-)
As a side note: Now that you can legally license the ROMs from Atari, and Oscar
has re-manufactored the Tempest spinner, and with us manufacturing the PC based
Vector generator, if you come up with a new X/Y monitor, a Tempest game could be
built from scratch using all new components! And no legal hassles! (Well maybe,
if you tried to put it on location, I'm not sure how the StarRoms license
handles that...)
Now that's cool! (Tempest being my favorite arcade game!)
Ok, back to your normally scheduled thread...

Post by juice28

Post by Zonn
I have no idea if any of the above will effect the stability of a direct

gnd

Post by Zonn
connection. Though it seems like without a direct gnd connection you are

not

Post by Zonn
running a DC amplifier, which (if this is truly the case), will cause

image

Post by Zonn
drift. If more vectors are drawn on one side of a screen than the other,

the

Post by Zonn
side with the larger amount of vectors will drift towards the center as

the

Post by Zonn
unmarked capacitor next to the 1k resistor attached to pin 2 charges. But

this

Post by Zonn
may be invalidated by the "self centering" circuitry, since I don't know

how

Post by Zonn
that works.

It never drifted, but I thought the same thing hence trying to get one or
the other inputs to ground.

Probably that magic "Self Centering" stuff! Maybe it's also trying to fight
your connection to ground. If the self centering was not exactly 0v (or your
supplies were not exactly symmetrical), it could be the IC sees this offset and
starts biasing itself to correct for it, and since it's shorted to ground all it
can do is fight the short, causing the heat. The LM379 might not be as
aggressive in it's self centering. But this is nothing but a *wild* guess!
It could also be going into oscillation, for some reason, causing it to really
heat up as well, you'd need a scope to figure out what exactly was going on.
-Zonn
--------------------------------------------------------
Zonn Moore
Zektor, LLC
www.zektor.com
Remove the ".AOL" from the email address to reply.

[Zonn]

Post by juice28
Zonn,
I didn't take it as you being harsh... Its all good.
I was just Lazy when I built this circuit. I built it more or less in a
night just to see if it was possible then I quickly lost interest (such is
the case on many things) when it worked and the theory was proved to myself
I moved on to other things.

I know that feeling well!!

Post by juice28
Maybe sometime I'll revisit it with a dual amp (like the LM4765), scope
and actually do the math to see whats up haha.
It might just be better to use a couple of tip31C's and tip32C's in a
circuit similair to the Atari's for a whole new deflection circuit. Do the
differential amp deal with feedback and all. It would probably have a better
slew rate as you could jack the voltage up using the tips. Of coarse it
wouldn't just be plug and play.

The problem with the Atari "discrete" circuit is it's low gain. The low gain
requires you to use large current monitoring resistors (10ohms), which means at
lot of wattage must be dissipated, requiring large non-inductive power resistors
that have a tendency to flame (at least on the original Sega monitors).

The high gain opamp allows low value resistors (0.1 ohm) which means much lower
power dissipation, less heat generated, better reliability.

I have a broken Vectrex (dot in the center kind of thing), and what I'd like to
do is remove the yoke and CRT, and rebuild the electronics around it (higher
supply voltage so that you can overscan it and run it much faster, better HV so
it doesn't bloom all over the place, etc), to give it a nice stable picture. As
is, the Vectrex X/Y monitor sucks compared to *real* X/Y monitors.

The plan is then to build a "to scale" 1/2 size Asteroids cabinet (9" almost
equals 19" / 2), and using a ITX computer and a ZVG, build a Vector multigame
into the cabinet. I think that would be cool!

My plan is to use a couple of these:

LM12CL http://www.national.com/ds.cgi/LM/LM12CL.pdf

or these:

OPA541AP http://www-s.ti.com/sc/ds/opa541.pdf

This second one could directly drive a WG monitor's yoke. A freund of mine came
up with a full replacement WG / Amplifone schematic using the OPA541. He's very
good engineer and made some nice improvements to the original design. I'll ask
him if it's ok to post the schematic.

Either way of them a way overkill for a Vectrex (they'd probably just get a
little warm running a Vectrex). But they are real OP amps (none of that "self
centering" stuff) and require a ground connection if you don't want them
drifting all over the place.

Post by juice28
I've rewound wyse yoke for use on the Vectrex and they work perfect, but
again alas I am a lazy bastard and don't want to go into the manual yoke
winding bussiness :-)

You sure? With those yokes I could go into Half-Pint Asteroid Multi-Vector game
production! ;-)

Post by juice28
Also I wanted at some point to make a little 9" or 13" color monitor for
the Vectrex and see if someone could map the other 2 colors somewhere in
memory and build the dac circuit for them. Then some 6809 software guy's
could make color games :)

I have one of those 13" Amplifone monitors and the resolution sucks. You'd
definitely would have to go with a higher res tube. The low res tube a 13" has
way too big of pixels.

Post by juice28
Someone make a 2 dac circuit and map it into someplace out of the way of
anything else so its compatiable and let me know. I'll put together a 9" or
13" color vector for it.

You could always add two more sample and hold circuits for the next two colors!
(Talk about an overworked DAC!)

-Zonn
--------------------------------------------------------
Zonn Moore
Zektor, LLC
www.zektor.com

Remove the ".AOL" from the email address to reply.

[Zonn]

On Thu, 16 Oct 2003 21:04:02 -0700, in msg

Post by Zonn

Post by juice28
Zonn,
I didn't take it as you being harsh... Its all good.
I was just Lazy when I built this circuit. I built it more or less in a
night just to see if it was possible then I quickly lost interest (such is
the case on many things) when it worked and the theory was proved to myself
I moved on to other things.

I know that feeling well!!

Post by juice28
Maybe sometime I'll revisit it with a dual amp (like the LM4765), scope
and actually do the math to see whats up haha.
It might just be better to use a couple of tip31C's and tip32C's in a
circuit similair to the Atari's for a whole new deflection circuit. Do the
differential amp deal with feedback and all. It would probably have a better
slew rate as you could jack the voltage up using the tips. Of coarse it
wouldn't just be plug and play.

The problem with the Atari "discrete" circuit is it's low gain. The low gain
requires you to use large current monitoring resistors (10ohms), which means at
lot of wattage must be dissipated, requiring large non-inductive power resistors
that have a tendency to flame (at least on the original Sega monitors).
The high gain opamp allows low value resistors (0.1 ohm) which means much lower
power dissipation, less heat generated, better reliability.
I have a broken Vectrex (dot in the center kind of thing), and what I'd like to
do is remove the yoke and CRT, and rebuild the electronics around it (higher
supply voltage so that you can overscan it and run it much faster, better HV so
it doesn't bloom all over the place, etc), to give it a nice stable picture. As
is, the Vectrex X/Y monitor sucks compared to *real* X/Y monitors.
The plan is then to build a "to scale" 1/2 size Asteroids cabinet (9" almost
equals 19" / 2), and using a ITX computer and a ZVG, build a Vector multigame
into the cabinet. I think that would be cool!
LM12CL http://www.national.com/ds.cgi/LM/LM12CL.pdf
OPA541AP http://www-s.ti.com/sc/ds/opa541.pdf
This second one could directly drive a WG monitor's yoke. A freund of mine came
up with a full replacement WG / Amplifone schematic using the OPA541. He's very
good engineer and made some nice improvements to the original design. I'll ask
him if it's ok to post the schematic.

Ok, to follow up my own post... ;-)

I just realized your replacement deflection board would drive a Vectrex yoke
*beautifully*! It's perfect!

I'd forgotten all about that! Hey I'm as lazy as the next guy! Op-amps
shmop-amps! Yours is finished! ;-)

The ZVG has circuitry for handling the linearity correction (blooming) that's
missing on the WG deflection boards (Atari did this as part of the AVG for WG
monitor based games).

I could crank up the voltage on the yoke for a very fast, nice and stable X/Y
display w/overscan. (From the picture it looks like you even kept the low
voltage regulators! The low voltage rectifiers is there, heck, all I need is a
transformer! Nice!)

And maybe modify one of the Cinematronics *gold* cans to output the lower HV
that the Vectrex uses (or how modifiable is your HV replacement as far as
lowering to for use on a Vectrex CRT? The Vectrex can use a bit higher HV as
well, the brighter lines go out of focus easily.)

hmm... I just realize I'm much farther a long on this project than I thought I
was!

-Zonn

(Yeah Fred, I think we can work out some kind of trade here... ;-)

--------------------------------------------------------
Zonn Moore
Zektor, LLC
www.zektor.com

Remove the ".AOL" from the email address to reply.

[juice28]

This post might be inappropriate. Click to display it.

[juice28]

Yeah Sorry to run away with your disscussion Cecil ;) Your welcome to come
back in and pitch in some comments.

I just took a few readings off my vectrex while it was running and this is
what I got.

Anode = 6KVH (seems low but I am not used to working with little B/W tubes
only 19" color tubes)
G2-G3 = 170VDC
G1 = -30VDC to +50VDC through a potiometer for brightness
Cathode = +50VDC
and the heater is a rectified 12V AC

So no problem on the 170VDC G2-G3 voltage, but the anode voltage produced
from my new HV is going to be Way to high. I treaked one down testing and it
was still around 9-10 KHV at the 170VDC.

The -30 and +50V for the brightness can't come from the new HV either.

I bet I could come up with a little HV unit for these pretty easy though
sometime.

It looks like one would just need 3 small buck boosts switchers or one
switcher with muliple outputs. One for -30, one for +50 and one for +170VDC.
Then you could use the 50 or 170 for the Flyback. I would go with the +50 as
you could then use a Wyse flyback.

Fred

Post by juice28
Zonn,
I don't have a vectrex schematic in front of me to know the voltages it
needs for the CRT gun and HV, but this HV unit I have uses a switching power
supply for the b+ and gun voltage. It can be adjusted anywhere from +24V
(supply) up to +180V. I am not sure how linear the HV output would be to
this. It might take a bit of tweeking to get the HV in the right range. I am
guessing the Vectrex gun voltages to be around +90V so I bet the HV output
would be around 11-12KVH. Again just a guess and I am not sure how well the
focus and screen would be. If you give me the Vectrex spec's I'll hook one
to my vectrex and see whats up... Shoot I guess I could just pull out my HV
probe and see where the original stands. Stay Tuned.
I also looked at those real OP amps you mentioned. I think that they
also are a neat all in one solution, but I thought it might be hard to get a
replacement for them in a couple of years as they are not really a main
stream device.
I am also going to send you a wyse yoke I modified for vectrex so you
can test it out and copy it if you want. Pretty easy mod. I doubt you have
any problem copying it. It just takes time. Wind it close then adjust it
with a cross hatch on the screen, then glue the windings back so they don't
move. I have also done this for amplifone and WG6100 yokes I have made, but
again it sucks and takes along time.
Yes the new deflection should work great on the vectrex.... haha should
be over kill to the max. You probably want to install the pots in the low
voltage regulator section to bring the voltage down. You will see the places
for them. You will also notice that it's basically your LV2000 circuit
(shameless thief I am ;) and you will probably have to tweek the resistors
that go across the yoke to get the best picture without ringing.
Fred

Post by Zonn
On Thu, 16 Oct 2003 21:04:02 -0700, in msg

On Fri, 17 Oct 2003 02:01:00 GMT, in msg

Post by juice28
Zonn,
I didn't take it as you being harsh... Its all good.
I was just Lazy when I built this circuit. I built it more or less

in a

Post by Zonn

Post by juice28
night just to see if it was possible then I quickly lost interest

(such

Post by juice28
is

Post by Zonn

Post by juice28
the case on many things) when it worked and the theory was proved to

myself

Post by Zonn

Post by juice28
I moved on to other things.

I know that feeling well!!

Post by juice28
Maybe sometime I'll revisit it with a dual amp (like the LM4765),

scope

Post by Zonn

Post by juice28
and actually do the math to see whats up haha.
It might just be better to use a couple of tip31C's and tip32C's

in

Post by juice28
a

Post by Zonn

Post by juice28
circuit similair to the Atari's for a whole new deflection circuit. Do

the

Post by Zonn

Post by juice28
differential amp deal with feedback and all. It would probably have a

better

Post by Zonn

Post by juice28
slew rate as you could jack the voltage up using the tips. Of coarse it
wouldn't just be plug and play.

The problem with the Atari "discrete" circuit is it's low gain. The

low

Post by juice28
gain

Post by Zonn

requires you to use large current monitoring resistors (10ohms), which

means at

Post by Zonn

lot of wattage must be dissipated, requiring large non-inductive power

resistors

Post by Zonn

that have a tendency to flame (at least on the original Sega monitors).
The high gain opamp allows low value resistors (0.1 ohm) which means

much

Post by juice28
lower

Post by Zonn

power dissipation, less heat generated, better reliability.
I have a broken Vectrex (dot in the center kind of thing), and what I'd

like to

Post by Zonn

do is remove the yoke and CRT, and rebuild the electronics around it

(higher

Post by Zonn

supply voltage so that you can overscan it and run it much faster,

better

Post by juice28
HV so

Post by Zonn

it doesn't bloom all over the place, etc), to give it a nice stable

picture. As

Post by Zonn

is, the Vectrex X/Y monitor sucks compared to *real* X/Y monitors.
The plan is then to build a "to scale" 1/2 size Asteroids cabinet (9"

almost

Post by Zonn

equals 19" / 2), and using a ITX computer and a ZVG, build a Vector

multigame

Post by Zonn

into the cabinet. I think that would be cool!
LM12CL http://www.national.com/ds.cgi/LM/LM12CL.pdf
OPA541AP http://www-s.ti.com/sc/ds/opa541.pdf
This second one could directly drive a WG monitor's yoke. A freund of

mine came

Post by Zonn

up with a full replacement WG / Amplifone schematic using the OPA541.

He's very

Post by Zonn

good engineer and made some nice improvements to the original design.

I'll ask

Post by Zonn

him if it's ok to post the schematic.

Ok, to follow up my own post... ;-)
I just realized your replacement deflection board would drive a Vectrex

yoke

Post by Zonn
*beautifully*! It's perfect!
I'd forgotten all about that! Hey I'm as lazy as the next guy! Op-amps
shmop-amps! Yours is finished! ;-)
The ZVG has circuitry for handling the linearity correction (blooming)

that's

Post by Zonn
missing on the WG deflection boards (Atari did this as part of the AVG

for

Post by juice28
WG

Post by Zonn
monitor based games).
I could crank up the voltage on the yoke for a very fast, nice and

stable

Post by juice28
X/Y

Post by Zonn
display w/overscan. (From the picture it looks like you even kept the low
voltage regulators! The low voltage rectifiers is there, heck, all I

need

Post by juice28
is a

Post by Zonn
transformer! Nice!)
And maybe modify one of the Cinematronics *gold* cans to output the

lower

Post by juice28
HV

Post by Zonn
that the Vectrex uses (or how modifiable is your HV replacement as far as
lowering to for use on a Vectrex CRT? The Vectrex can use a bit higher

HV

Post by juice28
as

Post by Zonn
well, the brighter lines go out of focus easily.)
hmm... I just realize I'm much farther a long on this project than I

thought I

Post by Zonn
was!
-Zonn
(Yeah Fred, I think we can work out some kind of trade here... ;-)
--------------------------------------------------------
Zonn Moore
Zektor, LLC
www.zektor.com
Remove the ".AOL" from the email address to reply.

[Zonn]

Post by juice28
Zonn,
I don't have a vectrex schematic in front of me to know the voltages it
needs for the CRT gun and HV, but this HV unit I have uses a switching power
supply for the b+ and gun voltage. It can be adjusted anywhere from +24V
(supply) up to +180V. I am not sure how linear the HV output would be to
this. It might take a bit of tweeking to get the HV in the right range. I am
guessing the Vectrex gun voltages to be around +90V so I bet the HV output
would be around 11-12KVH. Again just a guess and I am not sure how well the
focus and screen would be. If you give me the Vectrex spec's I'll hook one
to my vectrex and see whats up... Shoot I guess I could just pull out my HV
probe and see where the original stands. Stay Tuned.

From a simple glance at the schematic it looks like it uses 170 B+ voltages, +50
gun voltage, and I thought the HV was something like 9K, but a bit higher would
be fine.

Post by juice28
I also looked at those real OP amps you mentioned. I think that they
also are a neat all in one solution, but I thought it might be hard to get a
replacement for them in a couple of years as they are not really a main
stream device.

That by far is the best reason for using the discretes! Good point!

Post by juice28
I am also going to send you a wyse yoke I modified for vectrex so you
can test it out and copy it if you want. Pretty easy mod. I doubt you have
any problem copying it. It just takes time. Wind it close then adjust it
with a cross hatch on the screen, then glue the windings back so they don't
move. I have also done this for amplifone and WG6100 yokes I have made, but
again it sucks and takes along time.

Sure I'll look at, but don't even think I'm going to re-create it! ;-)

Post by juice28
Yes the new deflection should work great on the vectrex.... haha should
be over kill to the max. You probably want to install the pots in the low
voltage regulator section to bring the voltage down.

We'll see about lowering the voltages. The currents will be regulated by the
amplifiers feedback, and even if that fails, even assuming the yoke is 0ohms,
the feedback resistors of 10ohms will limit 25v to 2.5 amps, which I'm sure is
not going to burn up the yoke windings. Currently the Vectrex is being driven
with ridiculously low +/- 9v, and not enough current to even sustain the traces
at the very edges of the screen. Maybe I'll lower your voltages to about +/-
20v.

Post by juice28
You will see the places
for them. You will also notice that it's basically your LV2000 circuit
(shameless thief I am ;) and you will probably have to tweek the resistors
that go across the yoke to get the best picture without ringing.

I won't deny your thievery, just your victim! The LV2000 was designed by Anders
Knudsen. I had a design at the time that I was using for my own monitors, but
Anders design used the same emitter followers that Atari used for the pass
transistors, so his was a much easier drop in replacement than mine. His was a
nice design (both of our are basic textbook regulators), and if it's what's in
your deflection board, then your deflection board, then all the better!

-Zonn
--------------------------------------------------------
Zonn Moore
Zektor, LLC
www.zektor.com

Remove the ".AOL" from the email address to reply.

[Clay Cowgill]

Post by Zonn
OPA541AP http://www-s.ti.com/sc/ds/opa541.pdf

That is a neat part... I kept looking at the Apex power opamps as a nice
low-parts count solution, but they were just too proud of them ($$$). The
ST part in the ZIP package isn't bad!

Post by Zonn
A freund of mine came[...]

^^^^^

Was ist das, herr Moore? ;-)

-Clay

[Zonn]

Post by Clay Cowgill

Post by Zonn
OPA541AP http://www-s.ti.com/sc/ds/opa541.pdf

That is a neat part... I kept looking at the Apex power opamps as a nice
low-parts count solution, but they were just too proud of them ($$$). The
ST part in the ZIP package isn't bad!

Post by Zonn
A freund of mine came[...]

^^^^^
Was ist das, herr Moore? ;-)

Das spellink checkermeisterautocorrectingchoosendaswrongword!

;-)


What is it with German and the humongous words? I was looking at the side of my
new iron's packaging, and there was a paragraph of warnings about showering with
your plugged in iron, etc.

The German translation was like three 30 character words. It's like they had a
word for "the act of showering with your iron while it was plugged in".

Ah well, I've always envied those that can speak more than one language, I'm
still having problems with English...

-Zonn



--------------------------------------------------------
Zonn Moore
Zektor, LLC
www.zektor.com

Remove the ".AOL" from the email address to reply.

[Clay Cowgill]

Post by Zonn
What is it with German and the humongous words?

I vaguely remember some one-liner joke about that-- something to the effect
of "...a man was pronounced dead at the scene after he choked to death
trying to speak a 33 syllable German word..." ;-)

-Clay

[Cecil Casey]

I had better make myself clear. The first drawings were only of the pin
out's of the chips. If you thought they were a circuit you need to look at
them in a monospaced font. They were just to show that the same functions
existed on both chips, and to reference that they would both accept the same
drive and source voltages. I did not draw any circuit.

That said I will comment on the follow up posts forthwith...

Post by juice28
For some reason these newer amp
chips get really pissed when you run the inverting or non-inverting input
right to ground.

Since this drops into the Vectres Circuit you will see that it has both a
resistor and a bypass parallel bypass cap on the non-inverting path to
ground in the origional. I beleive they are R404 and C405.

Post by juice28
A quick glance shows that your running in a non-inverting mode whereas the
Vectrex used inverting amplifiers. Using your circuit as a direct

replacement

Post by juice28
would cause the images to be upside down and backwards, unless you also
reverse
the yoke connections

As I said that was just a pinout of both chips, and if you notice, both have
inverting, and non-inverting inputs to the op-amp. I was not suggesting
anything, just stating that was the pin out of each chip. I can see now
that I needed to make that clearer, Sorry :( I could have done better.

Post by juice28
You've also removed the linearity circuits which will cause the images to

bloom

Post by juice28
somewhat as they approach the edge of the screen.
Also missing is part of the spotkiller circuit which can lead to a small

burn

Post by juice28
mark in the center of the screen after too many power cycles.

Once again I was not clear I was posting pinouts of the two chips, not a
circuit to be droped in at that point. That said, since this chip can drop
in with a re-routing daughter board, the op-amp is following an r/c
intigrator with feedback, so as long as the new amp matches the curves of
the old amp you will have no problems. And I looked at both of the data
sheets and there is quite a nice match.

As for the Spot killer, it was driven by the horiziontal deflection, so
since this amp drops in the place of the old one, the killer works the same
as allways.

As for the R||C to ground see above.

I know that this reply may be a bit late, as it seems that Zonn has had some
sort of ephiniphy about vectors that I hardly understand. But does that
meen that we could order a free part and build a 'converter' board and fix a
lot of dead vectrexes?

I am still not clear. Not an unusual thing mind you.

-Cecil

[Zonn]

On Fri, 17 Oct 2003 06:00:41 GMT, in msg

Post by Cecil Casey
I had better make myself clear. The first drawings were only of the pin
out's of the chips. If you thought they were a circuit you need to look at
them in a monospaced font. They were just to show that the same functions
existed on both chips, and to reference that they would both accept the same
drive and source voltages. I did not draw any circuit.

Sorry Cecil, there's been a bit of a misunderstanding here.

Basically Fred and I hijacked your thread...sorry about that!

My comments were directed towards a schematic that was drawn up by Fred that was
pointed to by Clay at:

http://home.mchsi.com/~jstancliff/wsb/media/26501/site1060_t.jpg

This is the diagram with the non-inverting opamp configuration, etc. that I was
referring to...

Sorry about that!

-Zonn
--------------------------------------------------------
Zonn Moore
Zektor, LLC
www.zektor.com

Remove the ".AOL" from the email address to reply.

[Cecil Casey]

Post by Zonn
My comments were directed towards a schematic that was drawn up by Fred that was
http://home.mchsi.com/~jstancliff/wsb/media/26501/site1060_t.jpg
This is the diagram with the non-inverting opamp configuration, etc. that I was
referring to...
Sorry about that!
-Zonn

Hijack away! (My news feed ate that posting for some reason.) In your
discussions on using a replacement yoke I have a quick question.

Would updating the vectrex yoke make it easier to interface to arcade vector
video boards? I seem to remember a 2000 Christmas spoof of battlezone, and
the author noted that the Vectrex monitor was 'clipping' vectors in the Z
axis. I seem to remember him mentioning that vectors did not line up and
there was some curved distortion to the ends of the vectors. The higher
yoke voltages should correct at least the distorted vectors and improve the
sweep speeds right? Seems that this should not change the Vectrex greatly,
although you may have to dial down R401 and R408 to match the higher
deflection.

Please keep up the interesting posts!

[juice28]

Updating the Yoke won't have an effect, but updating the deflection with one
of these will http://www.arcadeshop.com/ (I am a bit biased on this product
though ;) ) This is actually a replacement for a WG6100, but like Zonn was
talking about it would increase the slew rate and let the vectrex work more
like a real vector monitor without foldback. You would only need a
transformer that put out around + and - 25 to +-30 volts AC to make this
work. This would make the Vectex monitor work alot better for the ZVG and
real arcade PCB's.

I am not sure about the Z axis problem you are speaking of. Maybe someone
can clear this up? Maybe the Vectrex uses a very slow phosphor or something?

Fred

Post by Cecil Casey

Post by Zonn
My comments were directed towards a schematic that was drawn up by Fred

that was

Post by Zonn
http://home.mchsi.com/~jstancliff/wsb/media/26501/site1060_t.jpg
This is the diagram with the non-inverting opamp configuration, etc.

that

Post by Cecil Casey
I was

Post by Zonn
referring to...
Sorry about that!
-Zonn

Hijack away! (My news feed ate that posting for some reason.) In your
discussions on using a replacement yoke I have a quick question.
Would updating the vectrex yoke make it easier to interface to arcade vector
video boards? I seem to remember a 2000 Christmas spoof of battlezone, and
the author noted that the Vectrex monitor was 'clipping' vectors in the Z
axis. I seem to remember him mentioning that vectors did not line up and
there was some curved distortion to the ends of the vectors. The higher
yoke voltages should correct at least the distorted vectors and improve the
sweep speeds right? Seems that this should not change the Vectrex greatly,
although you may have to dial down R401 and R408 to match the higher
deflection.
Please keep up the interesting posts!

[Zonn]

On Sat, 18 Oct 2003 11:52:05 GMT, in msg

Post by Cecil Casey

Post by Zonn
My comments were directed towards a schematic that was drawn up by Fred

that was

Post by Zonn
http://home.mchsi.com/~jstancliff/wsb/media/26501/site1060_t.jpg
This is the diagram with the non-inverting opamp configuration, etc. that

I was

Post by Zonn
referring to...
Sorry about that!
-Zonn

Hijack away! (My news feed ate that posting for some reason.) In your
discussions on using a replacement yoke I have a quick question.
Would updating the vectrex yoke make it easier to interface to arcade vector
video boards?

The yoke itself is just fine...

Post by Cecil Casey
I seem to remember a 2000 Christmas spoof of battlezone, and
the author noted that the Vectrex monitor was 'clipping' vectors in the Z
axis. I seem to remember him mentioning that vectors did not line up and
there was some curved distortion to the ends of the vectors.

This came up in r.g.v.a.c., Here's the original thread:

http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&oe=utf-8&frame=right&th=e0da7c0a22a106fa&seekm=pic7kvgikt87s8kh1mn2rv5q0tfdd84vfj%404ax.com#link1

or

http://tinyurl.com/rn1e

It's not the Z-axis that's at fault, but the delay between the Z-axis and the
X/Y axes.

Post by Cecil Casey
The higher
yoke voltages should correct at least the distorted vectors and improve the
sweep speeds right? Seems that this should not change the Vectrex greatly,
although you may have to dial down R401 and R408 to match the higher
deflection.

The higher voltages will indeed improve the Vectrex, by increasing its slew
rate, and decreasing the X/Y lag. I plan on using Fred's WG replacement
deflection board to run the Vectrex yoke directly, this should bring the Vectrex
CRT/Yoke combo up to an Arcade quality display. (Since it is smaller and the
trace doesn't have to move as far, it'll be faster. And since I plan on using
the ZVG to drive this, I can do edge bloom correction there, otherwise it would
have to be added if you wanted to use the display in the place of a real arcade
monitor like a G05. Atari didn't include linearity correction on the WG6100, so
of course Fred's replacement doesn't (can't!) do it either.)

The size of the display is not dependent upon the voltages used to drive the
yoke. The amplifiers feedback keeps everything in check, the larger voltage
allows the yoke to respond faster to changing inputs, but it doesn't change the
gain of the amplifiers, so R401 and R408 wouldn't (shouldn't!) change with the
change of power supply voltages.

-Zonn
--------------------------------------------------------
Zonn Moore
Zektor, LLC
www.zektor.com

Remove the ".AOL" from the email address to reply.

[o***@webtv.net]

Does this thread mean there is basically an upgrade available for the
Vectrex sometime soon?

[Clay Cowgill]

Post by Cecil Casey
I believe that National's newer part, the LM4765 will make a good
replacement for the LM379.

Fred did one using the LM1875 a while back...

Loading Image...

-Clay

[Richard Hutchinson]

Did a quick google of surplus suppliers -

http://www.cpcares.com/LM.html

http://www.stuutenbruin.com/linea/lineair10.html

http://www.jmb-electronique.com/INDEX%20Ampli%20audio.htm

http://www.techsonicusa.com/pages/827252/


I have a couple on standby, for when my Vec's start dieing :(




Richard H.