C64 9VAC any use?
va...@silvtrc.org
(0) +12V SID-6581 (+9V SID-8580)
(1) 6,8V for dassette motor (C3).
(2) 9V AC converted to a +5V positive pulse once per cycle to drive the TOD.
(3) 9V AC to the user port.
I'm just curious.. is there any "User port" peripheral that ever used this
mysterious 9VAC 100mA output ..?
The SID only uses the positive half of the AC. The dassette uses both positive
and negative cycles. The TOD only needs a positive pulse per cycle.
So if no userport peripheral minds. It should be possible to drive the 9V AC
input with an +12V square pulse that vary between 0V and +12V ..?
Dombo
> The 9VAC on Commodore 64 seems to be used for:
> (0) +12V SID-6581 (+9V SID-8580)
> (1) 6,8V for dassette motor (C3).
> (2) 9V AC converted to a +5V positive pulse once per cycle to drive the TOD.
> (3) 9V AC to the user port.
>
> I'm just curious.. is there any "User port" peripheral that ever used this
> mysterious 9VAC 100mA output ..?
Long time ago I made a RS-232 converter which used the 9VAC output to
convert the TTL levels on the user port to proper RS-232 levels. I can
imagine there are more user port devices that need more than 5 Volts for
their circuitry use the 9VAC output.
> The SID only uses the positive half of the AC. The dassette uses both positive
> and negative cycles. The TOD only needs a positive pulse per cycle.
>
> So if no userport peripheral minds. It should be possible to drive the 9V AC
> input with an +12V square pulse that vary between 0V and +12V ..?
Short version: it might work.
Longer version: the tricky bit is that there is bridge rectifier on the
9VAC line that rectifies both halves of the sine wave. With the solution
you propose there is only power about 50% of the time, so the capacitor
behind the bridge rectifier must bridge a much larger gap. Also with a
square wave there will be more current while charging the capacitor,
which means a bit more strain on the components, it more chance of
interference. If you could live without the TOD 12VDC should work fine.
va...@silvtrc.org
>> mysterious 9VAC 100mA output ..?
>Long time ago I made a RS-232 converter which used the 9VAC output to
>convert the TTL levels on the user port to proper RS-232 levels. I can
>imagine there are more user port devices that need more than 5 Volts for
>their circuitry use the 9VAC output.
The (+) side of things should manage. The (-) ouch ;)
Guess that's something to watch out for.
>> So if no userport peripheral minds. It should be possible to drive the 9V AC
>> input with an +12V square pulse that vary between 0V and +12V ..?
>Short version: it might work.
>Longer version: the tricky bit is that there is bridge rectifier on the
>9VAC line that rectifies both halves of the sine wave. With the solution
>you propose there is only power about 50% of the time, so the capacitor
>behind the bridge rectifier must bridge a much larger gap. Also with a
>square wave there will be more current while charging the capacitor,
>which means a bit more strain on the components, it more chance of
>interference. If you could live without the TOD 12VDC should work fine.
I have thought on this scenario too. But by having a 95% +12V .. 5% GND.
The capacitor won't have to supply power that long.
And the TOD only need a "blip" 50 or 60 times per second.
The bridge rectifier current might be problematic as you say. Question is if
it would be problematic enough. Esp, if the capacitor only need to supply
for 5% of the time.
An alternative is a H-bridge.
Linards Ticmanis
> The 9VAC on Commodore 64 seems to be used for:
> (0) +12V SID-6581 (+9V SID-8580)
> (1) 6,8V for dassette motor (C3).
> (2) 9V AC converted to a +5V positive pulse once per cycle to drive the TOD.
> (3) 9V AC to the user port.
>
> I'm just curious.. is there any "User port" peripheral that ever used this
> mysterious 9VAC 100mA output ..?
Yea, EPROM burners almost invariably use it to derive their various
programming voltages from it - usually 12.5V, 21V and 25V.
Also, I seem to remember that at least in the old "breadbox" C64, 9VAC
is used to create a "+5V CAN" voltage. This is fed to a number of things
that tend to create excessive noise on their +5V input line, so that the
"real" +5V line is not overly distorted. IIRC this includes the VIC.
--
Linards Ticmanis
Anssi Saari
> The 9VAC on Commodore 64 seems to be used for:
> (0) +12V SID-6581 (+9V SID-8580)
> (1) 6,8V for dassette motor (C3).
> (2) 9V AC converted to a +5V positive pulse once per cycle to drive the TOD.
> (3) 9V AC to the user port.
It depends on the board version, but if you look at
http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/c64/manual/c64-27r.gif
there is a rectifier for the 9 VAC and a charge pump which feeds 7812
and 7805 rectifiers. That 7805 output is marked "CAN +5" and goes at
least to VIC's VCC pin. So kinda important :)
http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/c64/manual/c64-26r.gif
va...@silvtrc.org
>http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/c64/manual/c64-27r.gif
>there is a rectifier for the 9 VAC and a charge pump which feeds 7812
>and 7805 rectifiers. That 7805 output is marked "CAN +5" and goes at
>least to VIC's VCC pin. So kinda important :)
>http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/c64/manual/c64-26r.gif
Checking the schematics it seems it would be a requirement that the voltage
shifts polarity to make the diode-capacitor charge pump to work.
So would the 9VAC input work with a +12V square wave ..?
The power supply part of C64 seems unneccesarily complicated.. ;)
Mr. X
news:4a115f2b$0$90264$1472...@news.sunsite.dk...
The CIA TOD input of 9VAC runs though a 74LS08 which is higher than spec to
begin with. I wouldn't put 12V pulses on mine.
Real RS-232 spec calls for a +/- 12V sender. Most of your modem adaptors
use the 9VC off the user port to build it.
The TODs could be fed off from DC off a 555 timer chip.
You could always check out Ben Heck's project. He started with a 64c board
and doesn't mention the 9VAC. I don't know if he just forgot to mention it
or if those parts don't work on his board.
http://benheck.com/04-05-2009/commodore-64-original-hardware-laptop
You haven't told us what you want it for. :)
If it's just a power supply, it may be easier to just build one from
scratch.
X
Anssi Saari
> Checking the schematics it seems it would be a requirement that the voltage
> shifts polarity to make the diode-capacitor charge pump to work.
>
> So would the 9VAC input work with a +12V square wave ..?
I guess not, but I really have no idea. I'd have to simulate with
Spice or something since analog electronics is hardly my strong
suit...
> The power supply part of C64 seems unneccesarily complicated.. ;)
Indeed :) Zimmers.net has also a 12 VDC mod somewhere. Basically you
just remove the 7812 and connect 12 VDC to it's output pad and feed
the 7805 from the same source and then connect the 7805 output to the
PSU +5V input. Puts quite a load on the 7805, though... For TOD, a
clock signal needs to be generated as well.
Anyways, it looks like things were simplified for the C64C, the same
+5 goes everywhere and the +9/+12 for SID is done from rectified 9VAC
input.
va...@silvtrc.org
>> AC
>> input with an +12V square pulse that vary between 0V and +12V ..?
>The CIA TOD input of 9VAC runs though a 74LS08 which is higher than spec to
>begin with. I wouldn't put 12V pulses on mine.
http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/c64/250469-rev.A-left.gif
If you check this schematic at C44, CR6, U23. You will see that the 9VAC is
connected via a current limiting resistor of 560 ohm to a 3V Zener.
This means that for the positive half the zener will prevent any voltage
above 3V. And thus as long as the zener+resistor can take it the voltage
will remain below 3V. Giving a positive "1" to U23.
For the negative half the zener will supply a limited current. That protects
the U23 from negative voltage and it will only see a "0".
U23 is an inverting schmitt trigger connected to both TOD inputs on the
6526 chips.
The maximum voltage to this circuit is dependent mainly on:
* Current capability of CR6.
* Heat dissipation limit of the resistor.
If only the heat dissipation of the resisitor is taken into account as the
current limit of zener CR6 is not specified. The maximum voltage at 9 VAC is:
Umax = Uz + sqrt(R * Pmax) = 3 + sqrt(560 * 0.25) = 14.8 V
Uz = Zener voltage (which lessens the voltage seen by the resistor)
R = Resistor used as current limiter.
Pmax = Maximum power dissipated over the resistor.
>You haven't told us what you want it for. :)
Replacement for the C64 power supply brick. Esp as an DC/DC converter that
can make use of battery power. But will not limit the functionality.
>If it's just a power supply, it may be easier to just build one from
>scratch.
That's what I'm proposing ;)
But need to have it clear what's the minimum specifications for such device.
Currently it seems a normal +5V DC and +12V "AC" square wave will do.
Mr. X
news:4a12cdfd$0$90275$1472...@news.sunsite.dk...
Of course, you're right. I completely missed the Zener. It does look like
it should work.
The print schematic I looked at is from the 1st ed 4th US print of
"Commodore 64 Programmer's Reference Guide" marked (c) Commodore 1982 looks
most like this one but has no R100 or C200.
http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/c64/manual/c64-32r.gif
> Replacement for the C64 power supply brick. Esp as an DC/DC converter that
> can make use of battery power. But will not limit the functionality.
>
>>If it's just a power supply, it may be easier to just build one from
>>scratch.
>
> That's what I'm proposing ;)
> But need to have it clear what's the minimum specifications for such
> device.
> Currently it seems a normal +5V DC and +12V "AC" square wave will do.
Cool. Switched?
X
va...@silvtrc.org
>it should work.
>The print schematic I looked at is from the 1st ed 4th US print of
>"Commodore 64 Programmer's Reference Guide" marked (c) Commodore 1982 looks
>most like this one but has no R100 or C200.
>http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/c64/manual/c64-32r.gif
One without R100 or C200 sounds it would lack the current limitation ..?
If it is that's yet another "gotcha!" :)
>> Replacement for the C64 power supply brick. Esp as an DC/DC converter that
>> can make use of battery power. But will not limit the functionality.
>>
>>>If it's just a power supply, it may be easier to just build one from
>>>scratch.
>>
>> That's what I'm proposing ;)
>> But need to have it clear what's the minimum specifications for such
>> device.
>> Currently it seems a normal +5V DC and +12V "AC" square wave will do.
>Cool. Switched?
Why not?, Nationals "simple switcher" series seems to make this way easier.
Buck-boost or Sepic are the most attractive types of conversion.
Especially this would eliminate any need for an explicit "C64 power supply".
Linards Ticmanis
> Replacement for the C64 power supply brick. Esp as an DC/DC converter that
> can make use of battery power. But will not limit the functionality.
In case you don't have success with your approach, here is a schematic
for a more traditional C64 PSU. It won't work on DC without a DC-AC
converter, but it's still a good bit stronger and more fault tolerant
than the original; it will simply blow one of its fuses rather than
frying your Commie when something goes wrong.
http://freenet-homepage.de/electronic-man/hardware/imag_h/netzteil.gif
"fl." means "flink"; German for "fast acting fuse". All the rest should
be clear. Use a 2A rated 7805.
--
Linards Ticmanis
Sam Gillett
"Linards Ticmanis" <ticm...@gmx.de> wrote ...
For what it is worth, here is another schematic for a heavy duty C64 power
supply:
http://personalpages.tds.net/~rcarlsen/images/c64ps.jpg
--
Best regards,
Sam Gillett
Change is inevitable,
except from vending machines!
Linards Ticmanis
That design is very similar, except it doesn't have the safety features
of the one I posted (fuses, zener-thyristor over-voltage discharger).
Not sure I'd trust my computers to it.
--
Linards Ticmanis
va...@silvtrc.org
>>> converter, but it's still a good bit stronger and more fault tolerant
>>> than the original; it will simply blow one of its fuses rather than
>>> frying your Commie when something goes wrong.
>>>
>>> http://freenet-homepage.de/electronic-man/hardware/imag_h/netzteil.gif
.
>> For what it is worth, here is another schematic for a heavy duty C64 power
>> supply:
>>
>> http://personalpages.tds.net/~rcarlsen/images/c64ps.jpg
>That design is very similar, except it doesn't have the safety features
>of the one I posted (fuses, zener-thyristor over-voltage discharger).
>Not sure I'd trust my computers to it.
I concur. However both lack any kind of protection on the 9VAC line.
A possible solution is a bridge rectifier that have the shown triac(?) on
the output.
Still think an 12V feed H-bridge 50/60 Hz square-wave will be able to
replace the 9VAC ;)
Sam Gillett
<va...@silvtrc.org> wrote ...
>
> I concur. However both lack any kind of protection on the 9VAC line.
> A possible solution is a bridge rectifier that have the shown triac(?) on
> the output.
> Still think an 12V feed H-bridge 50/60 Hz square-wave will be able to
> replace the 9VAC ;)
A much simpler solution might be an inexpensive and easily obtainable inline
fuse.
Could it be that I happened to think of that because by chance there are four
boxes of fuses sitting in front of my monitor?
--
Best regards,
Sam Gillett
Why is the third hand on a watch
called the second hand?
va...@silvtrc.org
>fuse.
Fuses are too slow.
Dombo
For the 9VAC line there is less need for a crowbar protection circuit
since there isn't a regulator that could go wrong.
The only way the 9VAC line could output too high voltage is if the
voltage on the primary side is too high. It might make sense to add a
surge protector on the primary side (fuse, resistor followed by a VDR).
A fuse on the 9VAC line might help to reduce damage to the power supply
itself and the attached computer in case there is something wrong with
the compute.
> Still think an 12V feed H-bridge 50/60 Hz square-wave will be able to
> replace the 9VAC ;)
Should work if the floating w.r.t. the 5V line. It might help to put a
coil in series to reduce current peaks and EMI when the H-bridge is
switching.
Clocky
A correctly rated fastblow on the 9VAC line is plenty fast IMO.
bud
Group: comp.sys.cbm Date: Sun, May 24, 2009, 7:01am (CDT+13) From:
nic...@migo.com (Clocky)
script:
>>Fuses are too slow.
>
>A correctly rated fastblow on the 9VAC
>line is plenty fast IMO.
Especially if the 9VAC is actually '12VAC' square-wave?
salaam,
dowcom
To e-mail me, add the character zero to "dowcom". i.e.:
dowcom(zero)(at)webtv(dot)net.
The fact that 'conventional wisdom' is indeed 'conventional',
does not, in any way, imply that it is wise.
Dombo
>
> Group: comp.sys.cbm Date: Sun, May 24, 2009, 7:01am (CDT+13) From:
> nic...@migo.com (Clocky)
> script:
>
>>> Fuses are too slow.
>> A correctly rated fastblow on the 9VAC
>> line is plenty fast IMO.
>
> Especially if the 9VAC is actually '12VAC' square-wave?
Not in this power supply.
Even if the 9VAC would be actually a 12V square wave it wouldn't matter
much. Mind you that the peak of the 9VAC sine wave is SQRT(2)*V = 12.7
Volts.
va...@silvtrc.org
Any transistor will protect the fuse in the event of failure. ;)
Clocky
> Group: comp.sys.cbm Date: Sun, May 24, 2009, 7:01am (CDT+13) From:
> nic...@migo.com (Clocky)
> script:
>
>>> Fuses are too slow.
>>
>> A correctly rated fastblow on the 9VAC
>> line is plenty fast IMO.
>
> Especially if the 9VAC is actually '12VAC' square-wave?
>
I wasn't commenting on the 12VAC square wave example.
bud
Group: comp.sys.cbm Date: Sun, May 24, 2009, 12:27pm (CDT+7) From:
do...@disposable.invalid (Dombo)
script:
>>Especially if the 9VAC is actually
>>'12VAC' square-wave?
>
>Not in this power supply.
>
>Even if the 9VAC would be actually a
>12V square wave it wouldn't matter
>much. Mind you that the peak of the
>9VAC sine wave is SQRT(2)*V = 12.7
>Volts.
Yes. First, I apologize to you and Clocky for not noting that the
comments were about _another_ PS. I was responding to the OP's comment
about the slowness of fuses. The OP has been advocating the use of
12VSqWAC for the 9VAC needs of a Commie.
I also know that fuses blow from over-current, not over-voltage. (Tho'
over-voltage is more likely to cause over-current.)
I am not an EE, so an EE may have to correct some of the following.
(The following quote is para-phrased.)
"The nominal voltage of an AC circuit is determined by the heat (energy)
dissipated by a resistance in the circuit. A sine-waveform voltage with
a peak of ~168 Volts will dissipate the same heat as 120 VDC."
In SqWAC, the voltage does not taper off as in sine wave, the current
reversal is (virtually) instantaneous. Therefore, 12V SqWAC has a
nominal voltage that is the same as its peak voltage.
IMO, 12 Volts in a 9 Volt circuit is not a good starting point, even
tho' 12 Volt square-wave has a peak that is slightly less than 9 Volt
sine-wave.
The OP also mentioned battery power. We might want to remember that the
lead-acid batteries used in autos have a peak voltage of _14_ volts.
Under the heavy loads in automotive service, they are rated at 12 volts
nominal.
If he uses other than an lead-acid battery, he shouldn't be concerned.
Dombo
You right about the near instant current switching, which is why I
suggested putting a coil in series in another post, though possibly the
coil on the 9VAC line in the C64 already does the job.
You are also right that the RMS voltage of a 12V square wave is higher
than a 9V sine wave. However in a C64 the 9VAC line is rectified and is
followed by a capacitor which means that after the diode you are left
with the peak voltage of whatever is at the 9VAC line minus the voltage
drop of the diode (typically about 0.7 volts). In other words the only
difference after rectification is that the voltage (and also the ripple)
would be slightly lower in the case of a 12V square wave even though the
RMS voltage is higher than in the case of a 9 Volt sine wave.
> IMO, 12 Volts in a 9 Volt circuit is not a good starting point, even
> tho' 12 Volt square-wave has a peak that is slightly less than 9 Volt
> sine-wave.
I don't see a problem with an H-bridge setup if it is floating w.r.t. to
the 5V ground.
> The OP also mentioned battery power. We might want to remember that the
> lead-acid batteries used in autos have a peak voltage of _14_ volts.
> Under the heavy loads in automotive service, they are rated at 12 volts
> nominal.
Well, if he wants to power his Commodore from a car battery he has much
more to worry about than than just the voltage (which would only be a
minor issue). In a car one has also to consider very nasty spikes which
without appropriate filtering would damage the C64. The easiest way (not
necessarily the most efficient way) to run a C64 in a car would be to
use an inverter and an ordinary power supply.