Part No. Description (Reference Designations in parentheses) ========== =========================================================== 24-250107 100 uf Aluminum Elec. Fixed Axial-Lead 25V Capacitor (C1,2) 31-1N914 75V Type-1N914 Switching Diode (CR2-4) 32-1N751A 5.1V 400mW Type-1N751A Zener Diode (CR1) 33-2N3906 PCB Switching and Amplifying Transistor (Q2, 4, 5) 34-2N3904 Type-2N3904 NPN 60V 1-Watt Transistor (Q1, 3) 52-222 22-Gauge Jumper Wire (2 in. required) 110000-223 22K Ohm, +/- 5% 1/4 W Resistor (R1, 8, 11) 110000-273 27K Ohm, +/- 5% 1/4 W Resistor (R5) 110000-393 39K Ohm, +/- 5% 1/4 W Resistor (R4) 110000-682 6.8K Ohm, +/- 5% 1/4 W Resistor (R6, 7) 110001-222 2.2K Ohm, +/- 5% 1/2 W Resistor (R10) 110001-472 4.7K Ohm, +/- 5% 1/2 W Resistor (R9) 110011-122 1.2K Ohm, +/- 1% 1/4 W Metal-Film Resistor (R2, 3) [110000-102 1K Ohm, +/- 5% 1/4 W Resistor (R')]
INPUT PROTECTION CIRCUIT PCB ASSEMBLY
[Figure 1, a scale picture of the top side of the PCB listing numbered connection points onto the deflection board, has been omitted for obvious reasons]
Y INPUT +27V X INPUT : : : : *J6 : : | : : \ : : CR1 / R1 : : 1N751A \ 22K : : 5.1V / : : \ | : : +--->|---+ : : | \ | : : R3 === | R2 : : 1.2K = | 1.2K : : 1% | 1% : J3*--/\/\---*--->|---+---|<---*---/\/\--*J1 | J4: CR3 CR2 :J2 | | : 1N914 1N914 : | / : : \ R5 \ To R601 1.6K 2% To R701 1.3K 2% / R4 27K / \ 39K \ + C2 - - C1 + / | 100 uf 100 uf | | 25 V J5 25 V | +-----|(-----+-----*-----+-----)|-----+ | | | | | | | === | | | R7 | = | R6 | +----/\/\----+ +----/\/\----+ | 6.8K +27V 6.8K | | : | | *J6 | | | | | \ | Q4 +-----+ Q3 R' / Q1 +-----+ Q2 2N3906| b |2N3904 1K \ 2N3904| b |2N3906 --- --- 1/4W/ --- --- c / \ e / \ c | c / \ e / \ c | | | +----------+----------+ | | | | | | | | | | | +--------------+--------------+ | | | | | | | | === \ / CR4 === | | = --- 1N914 = | | | | +--------------------+--------------------+ | Q5 | | 2N3906| b | --- | c / \ e | | +------------------+ | +---------+---------+ | | | | | | \ \ \ \ | / R8 / R11 / R10 / R9 | \ 22K \ 22K \ 2.2K \ 4.7K | / / / 1/2W / 1/2W | | | | | | J7*--------+ J9* *J8 *J5 : : : === -27V To D700 To D600 =
COLOR X-Y DISPLAY 92-053
INPUT PROTECTION CIRCUIT PCB
[Figure 13, an annotated picture of the deflection board PCB from the original TM-183 manual, has been omitted for obvious reasons.]
[Since you will be making the board from scratch and don't have the pictures that came with the board, I am providing the following text to describe the contact points on the deflection board. The following text is all mine and NOT Atari's but is derived directly from figures 1 and 13 that Atari did provide.]
All descriptions refer to a deflection board oriented component side up with the board label "P314" at the top (right-side-up) and the 2 large electrolytic capacitors at the bottom. This orientation will have the electrolytics "side by side" and the board will be "taller" than it is "wide".
SPECIAL NOTE: According to the Major Havoc manual, this board came in 2 versions. Later versions had a 1K Ohm, +/-5%, 1/4 W resistor as one of the "stilts" on the board instead of a plain piece of wire. If your board does not have this resistor, here is what you need to do.
I have included this resistor in my ASCII rendering of the schematic and labeled it R'.
[NOTE: Play Meter did copyright this article and the magazine is clearly marked with a copyright symbol.]
HOW TO MAKE A 'TEMPEST' MONITOR TROUBLE-FREE.
Atari's Tempest, when it is working, is not a bad game. It's just a shame that the monitors only last a few weeks (sometimes only days) between service calls. Most of the time, the monitor sits neglected on a techroom shelf.
Atari has developed a monitor protection board [included earlier in this text] to protect the monitor's components (2N3716 and 2N3792 "X" OUTput transistors, two each located on chassis frame) if and when there is a RAM lock-up on the Tempest CPU board. I want to clarify that the Wells-Gardner monitor is not at fault. Also note that on the later Atari games, the protection circuit has been incorporated into the board circuitry. [These statements seem to imply that this "fix" is compatible with the Atari upgraded P314s as well as P327s and P339s and will provide additional protection; if fact, I have seen it on a P327 before. It is sufficiently ambiguous that the exact opposite can be inferred. Judging from the areas of the board it alters, I would say it is incompatible (duplicates) the other fixes in this section.]
It's common knowledge that you can purchase a broken Tempest game cheap. With about 20 minutes of your time and a couple of dollars in parts, it is possible to add just six common components to the monitor deflection board and have a Tempest that will stay on location and work [Yeah, right]!
The parts required are:
ANODE +------+-+ CATHODE +--------+1N4737| +--------+ GROUND CATHODE +-+------+ ANODE / +------+-+ | --------+ | 1N914+--------+ === C700 R700 +-+------+ \ +--------+ | +--------+ 1K ohm +--------+ -27 VOLTS +--------+
Locate C700 in the top left of the monitor deflection board. (See Figure 13 on page 20 of Atari TM-183 Wells-Gardner Quadrascan service manual; second printing) and solder the cathode of the 1N4737 to the ground side of C700 (right side in Figure 13). Solder the end of the 1K OHM resistor to the -27 volt side (left side) of C700. Solder the one remaining wire (the cathode of 1N914) to the "X" INput side of R700 (top end of R700). You are now halfway finished.
ANODE +------+-+ CATHODE GROUND +--------+1N4737| +--------+ | +------+-+ \ CATHODE +-+------+ ANODE C701 === +--------+ | 1N914+-------- | +--------+ / +-+------+ R700 +27 VOLTS +--------+ 1K ohm +--------+ +--------+
Locate C701 (top middle in Figure 13) and solder the anode of 1N4737 to the ground side of C701 (right side). Solder the end of the 1K OHM resistor to the +27 volt side of C701 (left side). Go back to the same "X" INput side of R700 and solder the remaining wire (the anode of the 1N914) to this connection. Make sure that you have a good solder connection at the "X" INput of R700 as you now have a three-wire joint.
Cut the trace just to the left of pin 3 of P101 isolating the 5 rightmost connections (F700, R808, pin 5 of P701, C104, and R712) from the rest of the cluster. Jumper the 5 isolated connections to pin 1 of P100. This makes those 5 connections run at 33.8V instead of the usual 27.8V.
I am not sure what is gained by biasing these portions of the circuit by larger magnitude voltages but will look into it later.
When I saw this next hack, I couldn't believe my eyes; especially after I plugged it in and IT WORKED! The hack described above was implemented but with the following additions...
Evidently this operator got so tired of fixing the low voltage power supply regulator circuitry that he ELIMINATED IT FROM THE CIRCUIT! What he did was remove virtually that entire section (R100-103, D104-105, and P100) and instead formed the voltages required by dropping the unregulated +/-33 voltages across resistors! He connected a 10 Watt, 25 Ohm (+/- 10%) resistor between pins 3 and 7 of P100. An identical resistor was connected between pins 1 and 4 of P100. I'm not sure what kind of solder he used but it is takes a good minute for a 30 Watt soldering iron to even begin to melt a small portion of it.
This change will give you a very noisy approximation of the original voltages that should be good enough to run the monitor. Needless to say, the ceramic resistors get unbelievably hot and the screen is a little jumpy when drastic changes in drawing (current pull) occur (such as between waves and during the demo) but other than that the results are quite tolerable. If run for extended periods, this setup is virtually guaranteed to blow some fuses on the deflection board.
Since there is no longer a Q101 to worry about, it won't cause you any problems and since P100 no longer exists (either it or the connector that goes to it should be removed to avoid somebody plugging it in and adding the transistors to the now foreign circuit), there won't be any problems with Q102 and Q103 either (they are no longer required in the circuit and the connector that went on the now missing P100 just hangs in the air). If you are going to be placing one of these monitors out in the field or it is going to get frequent, extended duration use, this hack MIGHT be worth trying (assuming the degradation of picture quality is acceptable to the viewers) but I would think it would greatly stress the rest of the circuitry as well as the yoke coils and would limit the lifetime of the unit in other, less familiar ways. The PCB I saw this on had extensive burns on the amplification portions of the circuit which I almost never have to repair so BEWARE; this hack may have been the cause not the solution!
Go to [Table 'O Contents]