I'm exhausted, and a newb...

[philba]

In order to get it to this stage, I did what Len said at the 'EDIT' of your post, connected the '154 and resistors to +5v, and the hex chip Vcc to Z. But one major thing to note. I disconnected the wire which commons the ground of the new circuit to the ground 0v of the joystick, if thats connected nothings works at all, all y1-y6 are off. So how on earth is this new circuit getting power??? If the new circuit ground is not connected to the ground of the power source??

What is the reading between the 154's Vcc and Joystick ground? also, what is the reading between the 154's gnd and joystick gnd when seperated?

 

[ljcox]

YES! It works!... Well, 80% working....

Gears 1,2,3,4 work perfectly
Gears 5,6 seems to activate all 6 y1-y6 buttons, not sure why.

In order to get it to this stage, I did what Len said at the 'EDIT' of your post, connected the '154 and resistors to +5v, and the hex chip Vcc to Z. But one major thing to note. I disconnected the wire which commons the ground of the new circuit to the ground 0v of the joystick, if thats connected nothings works at all, all y1-y6 are off. So how on earth is this new circuit getting power??? If the new circuit ground is not connected to the ground of the power source?? Yes, this is very odd.
Any ideas on gears 5/6?

EDIT: Ok, I've tested the '154 inputs in each gear.
Negative probe on on isolated new circuit ground. And Positive probe on the entry side of each resistor. It seems the highs are 1.8v, They should be all +5V. and the lows are 0v. This is true for each gear, except 5,6, where the highs are 0.8 volts and the lows 0v. I suspect that 0.8v is not high enough to register a High on the '154 Yes, 0.8 is too low, thus all four inputs are low, which yields a combination of many results. Should I change the value of the resistors to a lower Ohm value? No, there is a serious problem somewhere, see below.

Thanks again for all your guys help, and thanks for the drawings Len, hopefully we dont have to use them though. You're welcome.

tnx!

It appears that with the Gnd open, the circuit is obtaining a connection to x via the outputs of the inverters (ie. the 5 that are low) through the pull down resistors in the JS to x. Hence the low voltages measured at the inputs to the 74HC154.

Our problem all along has been that we don't know what is inside the JS. We can only deduce what may be in it from your measurements and our general electronics knowledge.

The first step is to re-connect the Gnd of the circuit to the x wire. I know it does not work but we need to find out why. Remove the connection from pin 14 of IC2 (the inverter) and connect it to +5V.

Then connect a LED as per my LED option circuit from the appropriate output of IC2 to y1.

Test. If it works, connect the remaining LEDs.

If not, make voltage measurement on both sides of the LED with the output of IC2 High and then with it Low.

 

[ljcox]

Here is a variant that may solve the problem.

I suspect the problem with the last attempt was due the input protection diodes in IC2 injecting a current into Z.

So insert resistors R5 ~ 8 and transfer the wire on IC1 pin 24 from +5V to Z. BUT leave the top end of R1 ~ 4 connected to +5V.

The resistors can be any value in the range 47k to 100k and they don't need to be equal, so use what you have on hand. They are intended to minimise the current going into Z via the IC1 input protection diodes.

If it does not work, don't bushrange. Simply make a systematic measurement of the voltages for each gear and fill in a table as shown in the drawing so that Philba and I can analyse.

Also measure the voltage on Z for each gear. I forgot to add this to the table.

 

[Franknstein]

What is the reading between the 154's Vcc and Joystick ground? also, what is the reading between the 154's gnd and joystick gnd when seperated?

-Positive probe on '154 Vcc, Negative on joystick ground: 5.11V (new circuit ground not connected)

-If ground of new circuit is connected to joystick ground: 5.11V

-Between Vcc, and new circuit ground (not connected to joystick ground), 2.52v


LEN:

I have reconnected circuit ground to joystick ground, and powered the Hex Ic with 5+ from joystick as stated. I have connected each LED in series after the hex output and into y1-y6 inputs as per your diagram on pg4. I get similar results to previous test:

Gear1. just y1 led lights up
Gear2. just y2 led lights up
Gear3. just y3 led lights up
Gear4. just y4 led lights up
Gear5. y5 lights up as S3 is activated, (mechanically, S3 activated first, then S2, to make the combination)... as soon as S2 is switched too, y5 led light turns off, and y6 LED turns on!
Gear6. y6 does not light up, nothing happens at all.

VOltage test across LEDS:
Gear1: Engaged: -1.13V Disengaged: +3.10V
Gear2: Engaged: -1.17V Disengaged: +3.40V
Gear3: Engaged: -1.10V Disengaged: +2.70V
Gear4: Engaged: -1.15V Disengaged: +3.10V
Gear5: Engaged: +3.25v Disengaged: +3.29v
Gear6: Engaged: -0.02v Disengaged: +0.02v

Gotta do some other work now, see if I can do the other resistor circuit a little later this evening.

 

[ljcox]

I suggest that you do as I said in my previous post. If it does not work, produce a table of all voltages wrt Gnd.

I'll be at work tomorrow, so I won't be able to respond until about this time.

 

[ljcox]

LEN:

I have reconnected circuit ground to joystick ground, and powered the Hex Ic with 5+ from joystick as stated. I have connected each LED in series after the hex output and into y1-y6 inputs as per your diagram on pg4. I get similar results to previous test:

Gear1. just y1 led lights up
Gear2. just y2 led lights up
Gear3. just y3 led lights up
Gear4. just y4 led lights up
Gear5. y5 lights up as S3 is activated, (mechanically, S3 activated first, then S2, to make the combination)... as soon as S2 is switched too, y5 led light turns off, and y6 LED turns on!
Gear6. y6 does not light up, nothing happens at all.

VOltage test across LEDS:
Gear1: Engaged: -1.13V Disengaged: +3.10V
Gear2: Engaged: -1.17V Disengaged: +3.40V
Gear3: Engaged: -1.10V Disengaged: +2.70V
Gear4: Engaged: -1.15V Disengaged: +3.10V
Gear5: Engaged: +3.25v Disengaged: +3.29v
Gear6: Engaged: -0.02v Disengaged: +0.02v

The voltages across the LEDs are only of limited value.
It would help me if you measure the voltages wrt Gnd and tabulate them as suggested in the previous post.

 

[Franknstein]

Ok, now we're getting somewhere. I found 2 more bugs in the '154 outputs to the hex inverter (re: 5 and 6 gear). Now, all 6 gears turn on each LED individually and respectively. I tested voltages, and I get +5V to the LED when it activates it and 0v when not. So it seems the circuit is sending out active high signals to each gear input perfectly, I think its safe to say the new circuit logic is 100%. Now its just a case of interfacing to the js correctly.

I looked through my bin of stuff, and would you believe I have no resisters with values that high. Highest I had was 10k, so I will go pick some up tomorrow, and then do the table as you've described.

Tnx

Thought you might like to see what the whole setup looks like.
**broken link removed**
**broken link removed**

 

[ljcox]

Here is the image I could not post in the PM.

 

[ljcox]

Thanks for the photos. I knew that there was a fault or 2 around the g5/6 region, but I decided to let you find it. You'll learn more that way.

You did not say if the circuit works with the LEDs.

 

[Franknstein]

The circuit lights each LED individually, when each gear is selected, but does not activate anything on the pc joystick test utility, very strange.

But I'm gettting the resistors a little later today, will update with results late tonight.

tnx for the 4511 diagram. May I ask why you changed some of the diodes to resistors?

 

[ljcox]

Look up diode resistor gates. You need a resistor to act as a pull up or pull down. Pull down in your case. If you only use diodes, there won't be any change of voltage. When all diodes are off, the only pull down current will be due to the diode leakage which is often in the nA or pA range.

So when all diodes are off, with CMOS (which has a very high input resistance) the voltage at the inputs to the 4511 would depend on the diode leagage currents and the resistance of the PCB!

 

[ljcox]

Remember that diodes will drop about 0.6 Volt, so the 4511 won't see the full 5 Volt. This may not be critical, but I personally would prefer to use NAND gates to do the OR function. A triple NAND gate IC would do the trick.

 

[ljcox]

Here is a simple solution. I don't know why I did not think of it earlier. It is the closest you'll get to the original switches.

You will need 6 opto couplers such as the 4N25. Or, if you can find one with 2 opto in a package, you'll only need 3 ICs. Dual optos used to be available, but I could not see any in the electronics catalogues at a quick look.

 

[Franknstein]

That seems very logical. Thats what I was saying earlier in this post, about 'just closing a switch' like the original circuit, I wiki'ed the 4N25 and I understand the system perfectly, isolates voltages into two seperate circuits, and implements a remote switch using a light sensing device, very clever. That seems perfect! Will try pick some of them up today too.

EDIT: another thing I was thinking last night, is we could just use 6 of those relays I tried using originally on this circuit... same principle right?

Thanks again.

 

[ljcox]

You're welcome. Good Night

 

[Franknstein]

Another thing I was thinking last night, is we could just use 6 of those relays I tried using originally on this circuit... same principle right?

 

[ljcox]

Yes, you could use 6 relays but you can do it with 3 as I posted originally.

Here is a variant of that post. Now we know that Z has to be connected to the lever spring of R2.

 

[Franknstein]

Sorry guys, my internet line was down last night, so couldn't even build the circuit because I couldn't find your diagram online. Will see if its up tonight.

 

[Franknstein]

They finally fixed my internet connection, and I found some time this evening to work on the cicuit again...

FInally! It works 100%!
Thanks to all of you that helped out, definitely couldn't have done it without you all.

I used the 4N25's just as you drew the diagram Len, and it worked perfectly first time.

Next I want to add on to the circuit, with a 4511, and each gear will display the corresponding number on the 4511. Not sure when I will have the time for this next stage, will keep you posted though.

L8r
R

 

[ljcox]

That's good.

Here is the 7 segment circuit.

If you can't buy a 74HC148, the 74HC147 will do.

Note that the gear lines are connected in reverse order. This is because the outputs of the 75HC148 are active low. If you study the truth table in the data sheet you will see how it works. (1 and 6 are complementary as are 2 & 5, etc.)

If you can't buy the 74HC148 or the 74HC147, then buy a 4023 or 74HC10. These are triple 3 input NAND gates. I'' post a circuit for this option if necessary.