4 Digit Up/Down Counter IC?

[ppppenguin]

Originally Posted by jimmc101

What about recycling the bits from an old mouse?

That reminds me of a contraption I saw at Dorkbot meeting about 5 years ago. ISTR mice were used to measure the rotation of exercise bikes in a pedal controlled version of the "Pong" video game.

PS: Just found an online reference: http://www.lushprojects.com/cyclepong/

PPS: More on hacking mice as rotation sensors: http://lushprojects.com/microsoftmouse/

[Amraduk]

[Amraduk]

Hello Mark,

Originally Posted by mhennessy

We really need more information about this (Dave, that's a gentle hint )

You'll probably be pleased to hear that I've measured the speed of the spools during fast winding. It is 20mS per revolution, not surprising being as the motors are synchronous! So that's 50Hz maximum rotation speed.

Good point about the reset button...

Yes, a reset button is a must! Power-down wants to store the count so it can be restored on power-up. How often to update the stored count is something that will need to be decided. As long as it isn't lost in the case of a power-loss!

Regards,

Dave.

[Amraduk]

Hello Brian,

Thanks for your suggestion. I need a 4 digit display approx 30mm x 7mm so it will fit in place of the broken mechanical turns counter in my tape recorder. The ones you suggested are 8 digit and much too large to fit and are, therefore, unsuitable for my application.

Regards,

Dave.

[Amraduk]

Hello Mark,

Originally Posted by mhennessy

I said "it'll be" because I don't think it's been selected and installed yet.

You are correct!

Looking at the first post, the current device is mechanical - meaning Dave has some work to do - and my advice would be to choose the fewest cycles per revolution.

That was my intention.

In fact, I'd be tempted to DIY something, especially given the cost of quadrature encoders.

Yes, I was planning on doing it myself, finding a commercial product that could be fitted might be difficult.

A lot of old VCRs used these under the spool tables - they were reflective foil with black segments, and a pair of IR LEDs/phototransistors underneath - these could easily be modified for this application...

Again, they might be difficult to fit, it depends on how they are fitted and their size.

Pretty much the same applies to using parts from an old mouse! They generate many pulses per revolution, which would need dividing down. If I do it myself, I would arrange that only one pulse per revolution occurred.

Regards,

Dave.

[neon indicator]

A ball mouse has two quadrature encoders.

I've used them as panel controls remounting the spoked wheel between two bits of veroboard, one carrying IR emitter and one the dual o/p phototransistor. A PIC 18F will easily read them at full spooling speed without dividing and is much cheaper than an PIC16F84 (Comparison)

You can make miniature 7segment displays from 7 bar shape LEDs. Even Surface mount LED display can be made with regular iron on back of veroboard.

[ppppenguin]

Originally Posted by Amraduk

Pretty much the same applies to using parts from an old mouse! They generate many pulses per revolution, which would need dividing down. If I do it myself, I would arrange that only one pulse per revolution occurred.

If you look around the links on the "Cyclepong" pages that I cited you should find a lot of stuff about hacking mice. I remember one page I saw where the number of pulses per revolution was reduced by removing "teeth" from the optical interrupter. I haven't been able to find that page again but there is loads of stuff on the net about hacking mice.

[Amraduk]

Hello Jeffery,

Originally Posted by ppppenguin

I remember one page I saw where the number of pulses per revolution was reduced by removing "teeth" from the optical interrupter. I haven't been able to find that page again but there is loads of stuff on the net about hacking mice.

The problem I have is that the spooling motor has quite a large diameter shaft, so fitting on to it, one of the quite small discs from a mouse would, I think, be impractical. I rather think that there wouldn't be very much left of the disc once the hole had been enlarged enough to fit the motor shaft!

Regards,

Dave.

[neon indicator]

Use a drive belt?

[neon indicator]

Or
Just use the photodiode and LED from mouse and make one half of shaft reflective and the other half not where IR-LED/Phototransistor mounted. A rectangle of matt black paint.

[Amraduk]

Hello Michael,

Originally Posted by neon indicator

Use a drive belt?

You are joking, aren't you? However, all frivolity aside, that would mean additional work and expense, not to mention that rubber drive belts don't last forever, and I don't think it would be very practical.

Or
Just use the photodiode and LED from mouse and make one half of shaft reflective and the other half not where IR-LED/Phototransistor mounted. A rectangle of matt black paint.

I could use the LED and phototransistor from a mouse, though I'm not sure if I've got any old mice that I could, or even want to cannibalise! I wouldn't have thought it would be too expensive to buy. I do have an IR LED and photodiode I could use, though I'd have to buy at least one more of each for a quadrature output!

Painting half the shaft would be one way of doing it but I was thinking of using a disc so that the light from the LED would be completely cut-off and wouldn't depend on the reflectivity of the shaft surface, also, I'm not sure how well paint would adhere to the shaft, particularly at high speeds. There's also the possibility (probability?) that oil from the motor bearings could affect the paint.

Regards,

Dave.

[mhennessy]

Hi Dave,

Yes, I was planning on doing it myself, finding a commercial product that could be fitted might be difficult.

That's the spirit

Again, they might be difficult to fit, it depends on how they are fitted and their size.

A couple of pictures of the mechanics would be really interesting here

You'll probably be pleased to hear that I've measured the speed of the spools during fast winding. It is 20mS per revolution, not surprising being as the motors are synchronous! So that's 50Hz maximum rotation speed.

That's good. So the maximum count frequency is 50 time the number of segments/slots that your disc arrangement ends up with. Should be pretty easy to cope with

Power-down wants to store the count so it can be restored on power-up. How often to update the stored count is something that will need to be decided. As long as it isn't lost in the case of a power-loss!

There are numerous strategies for this, but the safest (IMHO) would be a little external circuity that detects the AC failing and causes an interrupt that saves the value to EEPROM. To make the best use of the charge in the smoothing capacitors, the display would be turned off during this, but once saved and verified, the thing could continue if the mains-fail input has gone back to normal - thus coping with "brown outs".

This scheme works well in my preamp, although I'd probably simplify it a bit (was trying to be clever and include low-voltage detect, which worked but probably wasn't needed). The simpler mains-fail in my A4 power amp is neater- only uses two transistors. I prefer discrete transistors in this role because their behaviour is much more predictable compared to ICs. No-doubt others will have their own preferences...

Cheers,

Mark

[KeithsTV]

Why not stick an alternating black/white strip around rotor of the take up spool motor and arrange two led/photodiodes to give a quadrature output from this strip.

Keith

[mhennessy]

Originally Posted by neon indicator

A PIC 18F will easily read them at full spooling speed without dividing and is much cheaper than an PIC16F84 (Comparison)

See post #70. I've only used the 16F84 because that's what is in our training boards. We've already said that we'll use the 16F827A or 16F828A to get some more IO to cope with power-down detection, etc.

[Amraduk]

Hello Keith,

Originally Posted by KeithsTV

Why not stick an alternating black/white strip around rotor of the take up spool motor and arrange two led/photodiodes to give a quadrature output from this strip.

Things that are 'stuck' on sometimes become 'unstuck'! I'd rather avoid that and use something that is more permanently attached, which would also allow the use of an opaque material that can interrupt the light, thereby cutting it off completely.

Regards,

Dave.

[ppppenguin]

How about a disc with alternate black/clear quadrants and a couple of those little slotted LED/photodiode devices.

[jimmc101]

Hi Mark
I think there's a typo, shouldn't that be the 16F627A or 16F628A?

Writing to the internal EEPROM is a relatively slow process (4 - 8 mS see table 17.1 of data sheet) so I think writing to it should only be done at power down.
As you have said advance warning of power down will be required to give enough time for the write to be carried out.

Jim

[Amraduk]

Hello Mark,

Originally Posted by mhennessy

A couple of pictures of the mechanics would be really interesting here

I don't have any photos but you can d/l a service manual from here: Revox G36 Service Manual. You should be able to see the arrangement in there.

So the maximum count frequency is 50 time the number of segments/slots that your disc arrangement ends up with. Should be pretty easy to cope with

Yes, I guess so. I had originally intended to have only one pulse per revolution but thinking about it, I realised that an approximately half-disc would unbalance the motor, so I'm going to make it two pulses per revolution, which will need dividing by two for the actual count.

There are numerous strategies for this, but the safest (IMHO) would be a little external circuity that detects the AC failing and causes an interrupt that saves the value to EEPROM. To make the best use of the charge in the smoothing capacitors, the display would be turned off during this, but once saved and verified, the thing could continue if the mains-fail input has gone back to normal - thus coping with "brown outs".

That seems like a good way to go!

Regards,

Dave.

[Amraduk]

Hello Jeffrey

Originally Posted by ppppenguin

How about a disc with alternate black/clear quadrants and a couple of those little slotted LED/photodiode devices.

That is more or less how I've decided to do it! I'm going to make it so that it generates two pulses per revolution rather than one so that the disc doesn't unbalance the motor! I'm not sure if that would be significant, but I think it would be better not to risk it.

One more post and we'll have reached the century!

Regards,

Dave.

[mhennessy]

That does look rather challenging!

I wonder if you could do something with the pulley that drives the original mechanical counter? Perhaps glue a slotted disc to it? Although there are some really good pictures in the manual, it's hard to get a sense of how much space there is when it's all assembled.

Presume you'll pick up the heater winding as a power supply?

Cheers,

Mark