HP-97 Printer Malfunction

[Jonas Sandstedt]

I have recently bought a HP97 calculator; it was in restored condition with re-built battery pack and new magnetic card driving wheel. When I received the calculator, it worked fine and printed perfectly, but the card reader would sometimes not read, and the seller told me he had had problems connecting the flat cable to the reader after repairing the drive wheel. I carefully disassembled the calculator (as described in several Internet places) and adjusted the flat cable. While doing this, the black wire connected to the printer feed button got loose, unnoticed by me. I re-assembled the calculator, and started it. Having removed the print paper, the printer tried to feed paper, but the printer head now moved erratically, first quickly and then very slowly when returning to the "home" position. I turned off the calculator, found the loose wire and re-soldered it. Now, the printer again tried to feed paper when the caclulator was started, but now seemed to move normally until it suddenly stopped in the "home" position. Since then, the printer head has not moved as it should anymore. I have manually fed paper into the printer by turning the motor forwards and backwards, but this has not changed the printer malfunction. When there is a printer action demanded, such as paper feed, nothing happens, but a faint whining sound is heard and the display reads "Error". When the calculator is turned off and on again, the sound is gone, and the calculator, including the card reader, works fine until a printer action is demanded. Then the above happens, and while the whining sound is heard, I have manually moved the print head by turning the motor by hand. It moves freely (the motor then not being energized), and the print head is moved 1-2 millimeters before the motor is turned on and the print head is quickly returned to the "home" position, and then the whining sound stops. So the motor works, and the print head is always returned when moved slightly from the “home” position. On the printer board, the transistors Q13 and Q14 measure OK, as well as the resistors R1 and R2. Pin 1 of U5 also measures OK (these should be checked according to the service manual). Now I turn to this forum for help, not knowing how to continue.

[ignazio]

Hi,
Have you checked the gears?

Ignazio

[Jonas Sandstedt]

I can advance paper manually by moving the gears at the motor, the printer head then moves and rotates the paper feed mechanism; I assume that this indicates that the gears are fine - but I might of course be wrong. Please let me know what gear errors I might check for in view of the above. I have not owned a hp-97 before, so this is new for me, only having reparied TI-59 calculators Before.

[Paul Berger (Canada)]

The gear the usually breaks is the middle idler gear, the normal failure mode is teeth get stripped off. If you need a replacement I have extras from a batch of aluminum gears another member had made recently.

You mention a faint whine, do you think this is the motor? It will display error if it detects that the print head has not moved.I would suggest that you take off the back and observe the printer when it attempts to print. You can either hold the battery pack in place, or do as I did, fashion a bar to hold it in place, I used coat hanger wire for this purpose, hooking one end under the board (be careful you don't short anything) and put the other end under one of the printers mounting screws. This will hold the battery in place so you may freely observe the printer.

The black wire on the forms feed switch being disconnected should not lead to erratic operation of the printer, it should only cause the forms feed button and end of forms switch to not work, so I would suspect there is something else wrong.

[Jonas Sandstedt]

Thanks for all advise. When the printer attempts to print nothing happens except the faint whining sound from somewhere in the printer assembly that starts. While the whining sound is there, I can manually turn the motor, so the motor is not energized at this stage. As soon as the printhead has been manually moved a small distance by my turning of the motor, the motor is energized, returning the printhead to the rest position, and the whining sound dispappears. The motor is thus not energized for forward motion, only for return motion. The seller said that the printer had worked OK for all years.

[Paul Berger (Canada)]

Hmmm Very odd, the only thing I can think of that would make a whirring noise in the printer would be the motor however you say it will not move in the forward direction. If the paper feed switch is incorrectly wired it can affect the forward motion, but only if you do not have paper in the printer. With paper in the printer the end of paper switch (OPS) would be open isolating the forms feed switch from the forward line. One thing that occurs to me is if one of the transistors that energize the motor in the forward direction is marginal and does not supply the motor enough power to move it forward but it moves easily when you give it a little push. You could try removing the middle idler gear and see if the motor moves under no load, also if you can hear a hum you may want to put your finger on the motor to feel for vibration. When you are manually moving it in the forward direction it probably energizes it in reverse when the home position switch drops out.

[ignazio]

Hi,
very strange behaviour. It's my opinion that you must check the home position switch too, disconnecting the two white/grey leads from the printer PCB and test the switch with an ohm-meter. When the head carriage is positioned near the right-hand wall , the ohm-meter should measure less than 1 Ohm and infinite when you rotatemanually the lead-screw away from the home position.
Test the cap connected in parallel to the motor leads too, and replace it if necessary.
Let me know the results of this check.

[Jonas Sandstedt]

Again thanks for all tips. I have checked the home switch, it reads 0.6-0.7 ohms in the home position and more or less infinite at the other end. I can’t measure capacitor values, but C2 is neither shorted, nor open-circuited. The motor runs very smoothly when powered manually by a finger, independently of if the calculator is powered or not, so I do not think there is anything that hinders its motion. When a printer action is required, and the whining tone is heard, the motor is powered with about 0.1 V DC. When the motor is manually moved such that the printhead is moved forward, the voltage is changed to about 5V for the properly functioning return motion. But I also noticed that when the motor is manually moved such that the printhead is moved backward (a very slight distance), the motor is powered in the forward direction, bringing the printhead back to the home position. As I interpret this, the motor may be electrically run in both direction, but in the home position it will not move. I cannot feel any vibration in the motor, but by listening carefully, it may be the motor that makes the sound.

[Paul Berger (Canada)]

Yes the motor is run electrically in both directions via Q13, Q14 and 4 of the transistors on U1. In the forward direction Q11 conducts through the base of Q9 which is in turn turned on and Q14 is is turned on through R1, connect Vb to one side of the motor and Q9 is the ground path. In the reverse direction it is Q10 that is the ground path and Q12 turns on Q13 through R2. Q8 is turned on to brake the motor by shorting the motor winding. The big difference between forward and reverse motion is in forward the motor speed is controlled by Pulse Width Modulating (PWM) the the signal on the forward line to base of Q11. I suppose a fault in the speed control circuit such that it always thought it was going to fast might could the pluses to be very narrow and not be enough to get the motor started.

It would be helpful to know where you measured this 0.1V. As far as I know the motor is just a PM motor so it would have a wound armature with a commutator and brushes as you turn the armature there may be null spots between the segments of the armature where there will be no continuity between the sides of the motor. Also note that the +5V and ground connections get reversed between forward and reverse. Since the forward signal is PWM measuring it with a meter will not give you a really good indication of the signal level. In between pulses the voltage generated by the coasting motor is monitored to determine how fast the printhead is moving. If the motor is stalled in the forward direction it might hum at the frequency of the PWM signal.

C2 is a 1uF non-polorized capacitor that would be there suppress the electrical noise from the motor, they could go bad, it probably would not hurt to disconnect one end to see if it makes a difference.

When I suggested taking out the idler gear I was not suggesting that there was a mechanical bind in the print mechanism, I am just thinking that the motor may be weakly driven in the forward direction and cannot get started under load.

[Jonas Sandstedt]

That was a really thorough explanation of the workings, great thanks. I only measured 0.1 V DC over the motor terminals, not knowing it was PWM in the forward direction. I have disconnected the plastic wheel after the motor in the wheel-train, and now the motor just continues to spin in the forward direction – you were right, the motor won’t get started under load, but starts without load. The motor does not stop spinning until the calculator is turned off, and the display shows “Error”. I have now also noticed that the motors makes a very faint start attempt under load, but stops accompanied of the whining sound. And yes, the sound probably comes from the motor, I can hear a similar note, but much fainter, when the motor spins freely without load.

[Jonas Sandstedt]

I have to make a correction: when the motor is manually moved in the forward direction and the power is set on, the motor returns the printhead to the home position. If I then press the button for advancing the paper, the motor runs a few turns in the forward direction, stops and the sound starts. If I then turn off the calculator and on again, and then press the button for advancing the paper, the motor does not run at all, only the sound starts. The motor thus only runs a few turns in the forward direction after having returned the printhead to the home position, not otherwise, when the printhead have been slightly moved due to the few motor turns in the forward direction there is no further movement.

[Jonas Sandstedt]

I know that I have not being able to provide any more information in this matter for a while, but have now, at last, been able to get some more electrical information. Getting no more useful data from my DMM, I have now borrowed an oscilloscope to find how the signals really look. When the motor is activated in the forward direction, the FWD signal, pin 5 of XA1P1, shows a square wave with an amplitude of 4V, each pulse having a duration of a little more than one millisecond and the period being a little less than 1.5 millisecond. The collector of Q11 has a DC component of 4V with 1V pulses on top, having a duration of about 0.5 millisecond and the period being a little less than 1.5 millisecond. The collector of Q9 shows the same square wave with a DC component of about 0.3V and the pulses having a total amplitude of 3.3V. The base of Q14 (the other side of R1) shows a similar pattern. When the motor is stalled and the tone is heard, the motor terminals both show a square wave with an amplitude of 3V, each pulse having a duration of a little more than one millisecond and the period being a little less than 1.5 millisecond. When the motor is released, a plastic middle gear being removed, a DC component of 2V is added to the square wave at the red motor terminal (the collector of Q14), the same square wave being present at the other motor terminal. The difference between the motor terminals at this moment is thus a DC voltage of about 2V. The control signal, pin 15 of XA1P1, is a square wave with an amplitude of 6V, each pulse having a duration of a little more than one millisecond and the period being a little less than 1.5 millisecond. The duration of the pulses changes slightly when the motor is released. I can see that the motor does not seem to receive the proper voltage, and it seems as if the problem probably starts at Q11 and/or Q9, but I am thankful for any kind of tips regarding what might be the error here.