A Tap Tempo Control Chip

[MoltenVoltage]

Steve at SmallBear said a bunch of you have bought our Universal Tap Tempo Control Chip MV-52.  Hopefully we will see plenty of designs for the contest (details on page 4 of this thread).

Deadline is May 31, 2009!   

[MoltenVoltage]

Beavis' insane Armageddon Processor, mentioned in another thread I just read, has a section called "Sequencer/VCF" that uses a 40106 to pulse a 4017 decade counter to sequence a voltage controlled filter:
http://www.beavisaudio.com/projects/ArmageddonProcessor/

Replacing the 40106 pulse circuit with MV-52 would be a simple way to give it tap tempo control.   

Note that you might need to add the 100 uF noise reduction cap shown on our application note:
http://www.moltenvoltage.com/downloads/Application_Note_1_for_MV-52_-_v1_1_-_Molten_Voltage.pdf

Here is a page full of VCF designs:
http://www.synthdiy.com/show/tags.asp?tag=vcf,%20voltage%20controlled%20filter

What might be even cooler is to run the pulse into 2 separate 4017 chips so you can sequence the frequency and resonance independently at each step.

[frequencycentral]

What might be even cooler is to run the pulse into 2 separate 4017 chips so you can sequence the frequency and resonance independently at each step.

That does sound cool, I've done it on my modular synth, but you only need to use one 4017 (or 4024). You can run each output of the 4017 to two (or three, or more) sets of pots for a multiple channel analogue sequencer. With the sequencers I've built this way I always use a buffer on each stage (can be opamp or transistor, two LM324 is my choice), with buffers you can drive many channels. I've never tried it without buffers but it may still work.

[moosapotamus]

I only just started tinkering with the MV-52, so maybe I'm doing something wrong, but I have a few questions about the way it is acting...

The pin output voltages are the same as the chip input voltage which can be anywhere from 2 - 5.5 volts DC.

I've got a 9VDC wall wart power supply and am using a 7805 regulator to to make +5VDC for the MV-52. So, based on the quote, above, I should be able to measure +5V peaks at pin 2 and at pin 3, right?
But at most, I'm only getting about 2.5V peaks at the on-off output (pin 2). And, I can't even measure the voltage output at the pulse output (pin 3), but it's not even enough to light an LED. I thought maybe since it's a very narrow trigger output, maybe the LED wouldn't have enough time to light up fully, so I tried plugging the pin 3 output into a PC based oscilloscope and that confirmed that the output of pin 3 is very low compared to the output of pin 2. Is that normal, or am I doing something wrong?

Another thing that I don't recall reading about is that there is about a 3 second delay when tapping in a new tempo. It seems to vary a little depending on if you are increasing or decreasing the tempo. But the tempo does not change immediately. As soon as I tap in a new tempo the output from the MV-52 just stops for a while, then starts with the new tempo after a couple of seconds. Again, is that normal, or am I doing something wrong here, too?

Thanks
~ Charlie

[MoltenVoltage]

I only just started tinkering with the MV-52, so maybe I'm doing something wrong, but I have a few questions about the way it is acting...

The pin output voltages are the same as the chip input voltage which can be anywhere from 2 - 5.5 volts DC.

I've got a 9VDC wall wart power supply and am using a 7805 regulator to to make +5VDC for the MV-52. So, based on the quote, above, I should be able to measure +5V peaks at pin 2 and at pin 3, right?
But at most, I'm only getting about 2.5V peaks at the on-off output (pin 2). And, I can't even measure the voltage output at the pulse output (pin 3), but it's not even enough to light an LED. I thought maybe since it's a very narrow trigger output, maybe the LED wouldn't have enough time to light up fully, so I tried plugging the pin 3 output into a PC based oscilloscope and that confirmed that the output of pin 3 is very low compared to the output of pin 2. Is that normal, or am I doing something wrong?

Another thing that I don't recall reading about is that there is about a 3 second delay when tapping in a new tempo. It seems to vary a little depending on if you are increasing or decreasing the tempo. But the tempo does not change immediately. As soon as I tap in a new tempo the output from the MV-52 just stops for a while, then starts with the new tempo after a couple of seconds. Again, is that normal, or am I doing something wrong here, too?

Thanks
~ Charlie

Hi Charlie,

We just re-tested MV-52 with an analog oscilloscope and a digital multimeter and confirmed that the output of pin 2 matches the input voltage for the chip on pin 1 (+5V DC using a 7805 voltage regulator and 9 volt wall wart), so we have to assume that you have set it up wrong.  The 0.1uF capacitor is essential for proper operation, and it must be connected as close as possible to pin 1.  Pin 4 must be held high with a 10K resistor for proper operation.

Are you using the exact setup on our datasheet:
http://www.moltenvoltage.com/downloads/Universal_Tap_Tempo_Control_Chip_MV-52_-_Datasheet_v2.pdf

There also shouldn't be any noticeable delay after pressing the second tap.  Are you only tapping twice?  Are you using a heavy duty pushbutton?  There is an upper limit on how fast MV-52 allows you to tap, which is .175 seconds.  If you are using a tactile switch, you can easily tap faster than that, but it's hard to tap any faster than that using a heavy-duty momentary on pushbutton like this:
http://www.smallbearelec.com/Detail.bok?no=672

Regarding the pin 3 output, we can easily flash an LED with it, provided it is connected without a resistor.  Granted, the pulse is extremely brief and the LED appears very dim, but if you use a high-efficiency clear red LED, you can easily see it.  The pulse width is around 1/20,000th of a second, making an accurate voltage reading difficult, but there is no reason it shouldn't be 5 volts, as it uses the exact same type of routine as pin 2.  In any event, the pin 3 pulse easily drives a 4017 decade counter (while pin 2 simultaneously drives an LED) as shown in our Application Note:
http://www.moltenvoltage.com/downloads/Application_Note_1_for_MV-52_-_v1_1_-_Molten_Voltage.pdf

Thank you for posting your questions.  Please let us know when you get it working and what you were doing wrong, because other people might have the same issues.  If you can't get it working, please post a photo of your breadboard setup and hopefully we can spot the problem.

[R.G.]

Just  curious.

How many people is "we" in Molten Voltage?

[MoltenVoltage]

Just  curious.

How many people is "we" in Molten Voltage?

Currently 4.  2 full time, 2 part time.

[R.G.]

Cool. Where are you located?

[MoltenVoltage]

Cool. Where are you located?

In the shadow of volcanoes:



Three Sisters, Mount Bachelor
Newberry Caldera, and Broken Top

http://en.wikipedia.org/wiki/Bend,_Oregon

The volcano on the front page of our website is South Sister (bulging about 10 feet a year)
The volcano on the logo is Mount Bachelor

[moosapotamus]

We just re-tested MV-52 with an analog oscilloscope and a digital multimeter and confirmed that the output of pin 2 matches the input voltage for the chip on pin 1 (+5V DC using a 7805 voltage regulator and 9 volt wall wart), so we have to assume that you have set it up wrong.  The 0.1uF capacitor is essential for proper operation, and it must be connected as close as possible to pin 1.  Pin 4 must be held high with a 10K resistor for proper operation.

Are you using the exact setup on our datasheet:
http://www.moltenvoltage.com/downloads/Universal_Tap_Tempo_Control_Chip_MV-52_-_Datasheet_v2.pdf

Hey, thanks for the reply. Yes, that is exactly how I have it set up.

There also shouldn't be any noticeable delay after pressing the second tap.  Are you only tapping twice?  Are you using a heavy duty pushbutton?  There is an upper limit on how fast MV-52 allows you to tap, which is .175 seconds.  If you are using a tactile switch, you can easily tap faster than that, but it's hard to tap any faster than that using a heavy-duty momentary on pushbutton like this:
http://www.smallbearelec.com/Detail.bok?no=672

Initially, I was just touching two wires together. But now I am using a small momentary push-button type switch, not a heavy duty stompswitch. I was also tapping several times, but since I read your reply I have only been tapping twice. I don't think I am taping faster than 0.175 seconds. But it does seem to make a difference how fast you tap. I have an LED on pin 2 and sometimes if I hold the first tap a little too long the LED starts blinking really fast. Then if I wait about two seconds and tap a second time the LED will then go out for several seconds and then start blinking really fast again. After trying that two second tap tempo a couple of times I can usually get it to work. But in general, it just seems like I have to be very careful about how long I hold each tap. Going from a slow tempo to a faster tempo also seems a lot easier than getting it to go from a fast tempo to a slow one. Slow to fast sometimes switches to the faster tempo relatively quickly. But there is almost always about a 3 second wait, while the LED is off, when going from a fast tempo to a slower tempo before the slower tempo begins.

I suppose there could be some intermittant problems with my breadboard (wouldn't be the first time that happened, right?). So I'll probably try moving it to a different spot, and maybe also try a different momentary switch. The way it's acting right now it's just hard for me to tell if really that sensitive to the duration of each tap, or if it's just a problem with my breadboard.

I suppose I should also mention that my point of reference for tap tempo pedals is a line 6 echo park. That's the only tap tempo effect I've really ever used and I think it works really well. I've had it on my pedalboard for a couple of years now, and like that it doesn't seem to care how many times you tap it. It usually takes me a few taps (while simultaneously playing the bass and not making any mistakes at the same time) to get the right tempo when either making a correction to stay in time with the drummer or to switch to a different tempo. Is it maybe too much to expect the MV-52 to perform equally well?

Regarding the pin 3 output, we can easily flash an LED with it, provided it is connected without a resistor.  Granted, the pulse is extremely brief and the LED appears very dim, but if you use a high-efficiency clear red LED, you can easily see it.  The pulse width is around 1/20,000th of a second, making an accurate voltage reading difficult, but there is no reason it shouldn't be 5 volts, as it uses the exact same type of routine as pin 2.  In any event, the pin 3 pulse easily drives a 4017 decade counter (while pin 2 simultaneously drives an LED) as shown in our Application Note:
http://www.moltenvoltage.com/downloads/Application_Note_1_for_MV-52_-_v1_1_-_Molten_Voltage.pdf

I switched from a red LED to a green one and I can see it faintly blinking now.

Thank you for posting your questions.  Please let us know when you get it working and what you were doing wrong, because other people might have the same issues.  If you can't get it working, please post a photo of your breadboard setup and hopefully we can spot the problem.

Will do, thanks.
~ Charlie

[MoltenVoltage]

Initially, I was just touching two wires together. But now I am using a small momentary push-button type switch, not a heavy duty stompswitch. I was also tapping several times, but since I read your reply I have only been tapping twice. I don't think I am taping faster than 0.175 seconds. But it does seem to make a difference how fast you tap. I have an LED on pin 2 and sometimes if I hold the first tap a little too long the LED starts blinking really fast. Then if I wait about two seconds and tap a second time the LED will then go out for several seconds and then start blinking really fast again. After trying that two second tap tempo a couple of times I can usually get it to work. But in general, it just seems like I have to be very careful about how long I hold each tap. Going from a slow tempo to a faster tempo also seems a lot easier than getting it to go from a fast tempo to a slow one. Slow to fast sometimes switches to the faster tempo relatively quickly. But there is almost always about a 3 second wait, while the LED is off, when going from a fast tempo to a slower tempo before the slower tempo begins.

Hi Charlie,

MV-52 was designed for 2 sharp taps, which is how we always tap when testing and playing.  Based on your experience, we discovered that if you hold down the pushbutton after the first tap more than .175 seconds, MV-52 automatically puts you at the top speed.

If you use 2 sharp taps, you won't have any problem.

The three second wait you occasionally experience most likely comes from holding down the button way too long (more than .5 seconds) which puts the chip back in the button press routine where it waits about 3 seconds for another tap.

You can input a new tempo by tapping twice at any time while the chip is in playback mode.  It doesn't matter whether you are going from a slow tempo to a fast one or the other way around, there won't be any noticeable delay between the second tap and the start of playback.

[slacker]

I've found exactly the same thing as Charlie, you need very precise taps to get it to work reliably. I think in my case the 3 second wait is caused by the cheap momentary push switch I'm using, presumably the chip either isn't reading both taps so it waits for the 3 second before "timing out" or it's switch bounce and it's reading more than 2 taps and again "timing out".
If a better switch doesn't solve the problems I'll probably just add a simple pulse generator/switch debouncer between the switch and the chip to guarantee a clean thin pulse.

[kvb]

add a simple pulse generator . . . to guarantee a clean thin pulse.

I was thinking the same thing.

the only pulse generator I have been able to find is on the MFOS sequencer; it uses a 40106

can a single transistor generate a pulse like that or does it have to be the 40106?

[moosapotamus]

All right. Thanks for the advise, Karl. And thanks for the confirmation, Slacker. I'm using a pretty cheap pushbutton switch, too. I didn't think that the action and timing of using my finger instead of my foot would make much difference. But if each tap must be between 0.175 and 0.5 seconds, I can see that timing is pretty critical. Seems like a pulse generator would be a good idea.

Just out of curiosity, what goes on in the chip if you tap more than twice?

I'm sure there are other options besides a 40106 for making a pulse generator, but I would not expect a single transistor to be able to do it. However, even though I only have very little experience programming uC chips, I would think that a pulse generator and a switch debouncer could both be done in the code, no?

Thanks
~ Charlie

[earthtonesaudio]

You can make a pulse out of a square wave easily with just a capacitor and a resistor or two:

http://sybarite.us/puertorico/wp-content/uploads/2008/06/edgetriggeringa555.gif

[moosapotamus]

Well, that sure does look pretty simple. I'm assuming that the cap sets the pulse width? If so, I wonder what value would be needed to ensure that it is between 0.175 and 0.5 seconds?

Thanks
~ Charlie

[earthtonesaudio]

Er... I was a little tired when I posted that.  The picture I linked to is just a passive differentiator.  It needs something on the output to make a clean pulse, such as a comparator, gate, schmitt trigger, 555 timer, or possibly just a transistor. 

As for calculating the part values, once you know your voltages, that's a nice application of the RC time constant. 

[MoltenVoltage]

Pushbutton switches, especially the heavy-duty kind, bounce significantly both when pressed and when released.  We repeatedly tested the standard heavy-duty momentary switches (again using sharp taps) and found that the outside limit of how long the switch would bounce (including both press and release) was .175 seconds.  As such, we used a hard debounce routine which simply waits .175 seconds after the initial press to continue on in the tap routine.  We wanted to make MV-52 work without a physical debounce circuit, which it does if you tap sharply, but it sounds like not everyone taps like we do.

Currently the first tap needs to be less than .175 seconds.  Any more than that and the chip will switch to top speed.  Any more than .5 seconds and it will jump back to the tap routine and pause for about 4 seconds.

We are going to revise the code and test the chip over the next couple days to deal with long button presses.  Will let you know as soon as we are done.

The chips can be easily reprogrammed, so anyone who wants the revised code can send their chip in for re-programming.

For everyone else, we have asked SmallBear to send us their remaining inventory so we can re-program the chips.  We will let you know as soon as the chip is available again, which should be within a week or so.

Just to be clear, the chip works as advertised if you tap sharply with your foot using a heavy duty button.  You don't need to send in your chip, but if you want to, the address is:

Molten Voltage
P.O. Box 3753
Sunriver, OR  97707

Be SURE to include your name and return address.  Also, be SURE to put the chip pins in hard foam or something to protect them from being crushed.

[nelson]

Pushbutton switches, especially the heavy-duty kind, bounce significantly both when pressed and when released.  We repeatedly tested the standard heavy-duty momentary switches (again using sharp taps) and found that the outside limit of how long the switch would bounce (including both press and release) was .175 seconds.  As such, we used a hard debounce routine which simply waits .175 seconds after the initial press to continue on in the tap routine.  We wanted to make MV-52 work without a physical debounce circuit, which it does if you tap sharply, but it sounds like not everyone taps like we do.

Currently the first tap needs to be less than .175 seconds.  Any more than that and the chip will switch to top speed.  Any more than .5 seconds and it will jump back to the tap routine and pause for about 4 seconds.

We are going to revise the code and test the chip over the next couple days to deal with long button presses.  Will let you know as soon as we are done.

The chips can be easily reprogrammed, so anyone who wants the revised code can send their chip in for re-programming.

For everyone else, we have asked SmallBear to send us their remaining inventory so we can re-program the chips.  We will let you know as soon as the chip is available again, which should be within a week or so.

Just to be clear, the chip works as advertised if you tap sharply with your foot using a heavy duty button.  You don't need to send in your chip, but if you want to, the address is:

Molten Voltage
P.O. Box 3753
Sunriver, OR  97707

Be SURE to include your name and return address.  Also, be SURE to put the chip pins in hard foam or something to protect them from being crushed.

That's gotta suck.

What about folks that the price of postage makes this rather pointless?

[MoltenVoltage]

What about folks that the price of postage makes this rather pointless?

Good point nelson.

If you send us proof of your purchase from SmallBear, we'll send you replacement chips.

Please email us directly rather than posting any information here.  Our email address is:

questions@MoltenVoltage.com