Frequency to voltage IC for analog delay rate indicator?

[armdnrdy]

I've been doing some research on designing a multi BBD test unit that will (hopefully) identify the delay time of the IC as well.

As I was trying to figure out the delay time part, I came across  frequency to voltage ICs.

After breezing through the data sheet, it occurred to me that an IC of this type just might be able to be incorporated into the wet signal of an analog delay and be used to light an led as a rate indicator.

I'm sure there are many choices out there, but I was looking at the LM2907.

The data sheet states: 

50 mA sink or source to operate relays, solenoids, meters,
or LEDs
■ Frequency doubling for low ripple
■ Tachometer has built-in hysteresis with either differential
input or ground referenced input
■ Built-in zener on LM2917
■ ±0.3% linearity typical
■ Ground referenced tachometer is fully protected from
damage due to swings above VCC and below ground
Applications
■ Over/under speed sensing
■ Frequency to voltage conversion (tachometer)
■ Speedometers
■ Breaker point dwell meters
■ Hand-held tachometer
■ Speed governors
■ Cruise control
■ Automotive door lock control
■ Clutch control
■ Horn control
■ Touch or sound switches

Okay, now who's going to design this? With my limited knowledge, it will be ready to go this time next year!!!

[Seljer]

I'm not sure what you're getting at with the LED as a rate indicator. Such a chip would be used to take a train of pulses (e.g. the clock signal for the BBD) and represent it as an analog voltage/current.

So you could use an analog panel meter type indicator for delay time which might be kind of cool 

[armdnrdy]

If my thinking is correct, the wet output of a delay is also a pulse of signals that happen at a certain frequency depending on the setting of the rate/delay time control.

Now if that pulse of voltage can be turned into something usable that would fire an LED, (maybe rectifying the signal) that would be a rate indicator.

[armdnrdy]

I was thinking about a panel type meter using the signal from the clock for the BBD tester. Unfortunatly, the signal from the CP pins does not give you the signal delay time from the chip.

I believe that a signal would have to be inserted at the input of the BBD, and then the signal from the output of the BBD converted by some means to something readable.

With no signal input you can take a frequency reading at the clock pins but if you try to take a  reading at the output of a BBD with no input signal you don't get a reading.


If anyone wants to chime in and give me some direction, I'll be glad to do the heavy lifting to make a BBD tester a reality!!

[Mark Hammer]

In the grand scheme of things, it might be simpler to use a course/fine system like analog synth oscillators use, or something equivalent to an octave switch on a synth, such that a 4, 6, or even 12-position rotary switch selects range, and a pot selects within the range.  You wouldn't be able to see a 3 or 4-digit number on a display, but if the goal is simply replicability of settings, then a range switch with defined times (and many circuits will show equivalents between timing resistors or caps and clock frequency) would accomplish a great deal.

Of course, if there is a need to see a numerical display, or see it from a distance, then I understand the need to adopt a more complex approach.

For more precise indication, one would probably want to opt for a microcontroller, PIC, or whatnot; ideally something that could read a pot value to establish clock rate, and both drive the clock via digital outputs, and also drive an LED or LCD display, corresponding to pot setting.

[armdnrdy]

Correct me if I'm wrong...

Let's go back to the bucket brigade scenario,

The AC guitar signal enters the BBD IC. The signal gets passed from bucket (stage) to bucket until it gets dropped out of the BBD output.

Now the "buckets" are passing the signal at a certain interval (frequency) but as I noted earlier, if you try to measure the signal at the output of the BBD, you are actually getting a frequency measurement of the AC guitar signal, 60Htz.

With that being said, the signal does drop out of the BBD at a certain interval and this timing does vary with BBDs of different stages, so what can we incorporate to count the number of times that it drops out?

I was thinking that there has to be some sort of logic/counting IC that can be used for this purpose.

Any ideas?

[Mark Hammer]

You're losing me here, Larry.  Maybe I'm being distracted by the subject heading. Is the goal here to have some visual indication of how much time passes between entry into, and exit from, the delay chip?  Or is it some other goal?

[maartendh]

With that being said, the signal does drop out of the BBD at a certain interval and this timing does vary with BBDs of different stages, so what can we incorporate to count the number of times that it drops out?

Even if you could do this, it would only work on the first note you play; once you play a series of notes it would not be possible to match the original and the corresponding delayed note, so the numbers of your indicator would be all over the place.

Clockspeed would give a good indication of the delay, just as the position knob of the speedpot will give you an indication of the delay to expect.  If you absolutely must, you can find the delay time in milliseconds by dividing the clockspeed you measure into half the number of stages of the BBD in question. So a 20 KHz clock into a MN3004 (=512 stages) gives 256/20,000 = 12,8 millisec delay; set the clock to 100 KHz and the delay will be 2,56 millisecs. Now all you have to do is to find a circuit that will do this for you, which probably is possible, be it a bit complex.

Would be much easier to measure clock speed at certain positions of your speedknob, calculate the corresponding delay time and mark this next to the knob, I think.

Maarten

[armdnrdy]

You're losing me here, Larry.  Maybe I'm being distracted by the subject heading. Is the goal here to have some visual indication of how much time passes between entry into, and exit from, the delay chip?  Or is it some other goal?

Hey Mark,

Yeah,
Two potential goals got merged into one.
The subject heading is about trying to find a way to add a rate indicator to an analog delay using some type of frequency to voltage IC.

The second issue is trying to find a way to identify the amount of delay leaving the BBD to incorporate in a universal BBD tester.

Last night I was looking into counting ICs. I need a way to count the milliseconds between pulses of the output of the BBD, and then display that  number on an analog display. Still a lot of research to be done. I know there's a way to do it, but I just don't know how just yet!

Since the max signal delay time varies between BBD ICs, this reading can be used to identify the authenticity of the BBD in question. I feel that with all of the counterfeit chips running around these days, a BBD tester that shows A.  that the BBD passes signal and B. that the max delay time is in the ball park would be very useful.

[Mark Hammer]

trying to find a way to identify the amount of delay leaving the BBD to incorporate in a universal BBD tester.

Last night I was looking into counting ICs. I need a way to count the milliseconds between pulses of the output of the BBD, and then display that  number on an analog display. Still a lot of research to be done. I know there's a way to do it, but I just don't know how just yet!

Since the max signal delay time varies between BBD ICs, this reading can be used to identify the authenticity of the BBD in question. I feel that with all of the counterfeit chips running around these days, a BBD tester that shows A.  that the BBD passes signal and B. that the max delay time is in the ball park would be very useful.

Okay, much clearer now.  And what the supplemental info indicates is that merely knowing how many clock cycles have passed, or pulses are being sent per second, does not provide an accurate indication of delay time if the chip itself is misrepresented as having X number of stages, but actually has Y.

I don't know that identifying the true capacity of a BBD necessarily has to be co-opted into a usable indicator of assumed delay time for a pedal.  Using a scope to send a predetermined pulse to a BBD, and seeing how long it takes to "get to the other side" (where undoubtedly, a chicken is waiting) is a plausible strategy.  And, once you've identified the capacity of a chip, you don't have to do it again, whereas a means for fine calibration of delay time in a pedal is something one would do many times over, even after knowng the capacity of the chip.

My vote goes to separate strategies for each objective, rather than trying to shoehorn them both into one.

[Fender3D]

Hi Larry,

you should define "rate indicator" better:

for the BBD test bed will be a hard job looking for precise measuring.
I mean, the easier way might be having a "switch" enabling a short signal to BBD, an input detector (maybe a simple overload indicator is enough), an output detector and a counter started and stopped by detectors.
the hard part is that every circuit involved will have a proper lag or response time, which will affect reading.
As I told you, it's faster with a new digital scope feeded with a signal burst, measuring A/B delay.

For a pedal kind of stuff,
if you think at it as the LFO rate mod available for phasers or vibing stuff, with LED blink each "delay cicle", then you just need to divide the clock frequency to a more "viewable" behaviour (ie 1 blink each 50-100 mS).

[armdnrdy]

Hey Federico,

For the test bed:
The "short signal burst" switch is on the right track. It would be like a runner leaving the starting line, and at that time starting a stop watch. When the runner reaches the finish line, you stop the stop watch. I just have to reproduce this scenario with a circuit.

I want to make a stand alone universal BBD tester because I know there are a lot of DIYers that do not own or have access to a scope.

My third build was an AD-80. I didn't have a scope yet and I purchased my MN3005 from a seller on ebay. (okay stop laughing)
The first chip I got kept overheating. The second chip from a different seller passed signal, gave delay, but when A/B'd against a carbon copy that I own (set at half delay rate) didn't produce the proper delay length. (re-label MN3008) Third time was a charm!!

Also, there are times when I source BBD's but don't have a finished build to test them in.

I think that it would save time, and a lot of frustration to have a stand alone BBD tester for many reasons.

[slacker]

Simple tester, put the chip in a delay pedal and set it to 3 or four repeats, plug the output into your computer and use something like audacity to record it. Play a staccato chord stab or a single click from a drum machine or metronome into it, you can then view the recorded output on your computer and measure the delay time from that.

This assumes you know what clock speed the delay pedal is running the BBD at, if not you'd have to build a simple delay circuit that ran the chip at known clock speeds so you would know what delay time to expect.

Another way is to play a stream of pulses into the delay at the same speed as the expected delay time, say every 300ms or whatever. If the delay time is exactly what you think it should be, then you won't hear the repeats as they will match up exactly with the dry signal, in reality you'll hear a flanging effect. If the delay time is significantly different to what you expect you'll hear rhythmic repeats. I've used this method to check the accuracy of tap tempo delays, play the drum machine through it, tap in time with the drum machine.  

[armdnrdy]

Slacker:

SIMPLE TESTER?

Thanks for the input, but that sort of testing is what I envision getting away from!

What I want is plug and play. plug the BBD into a socket on the test bed, push a momentary button, look at a readout located on the test bed. DONE! I believe this could become a reality with a fairly low count of components.

Fender3D (Federico) has a scaled down version of the generic Panasonic/Matsushita test circuit consisting of 3 capacitors, 4 resistors and a pot.

Add in a clock chip, an IC to generate a signal, an IC to count the signal, and one for the display. I can't imagine this circuit being that complicated. I think that a BBD tester of this sort would be an invaluable tool for someone who builds many different effects utilizing BBDs.

[slacker]

Ok then, here's a simpler way, build your generic test circuit with your clock. Build a simple pulse generator, this could be done with an opamps, a 555 timer or some other simple logic chips. Make this pulse at the delay times you want to test, connect it to a LED so you've got something that flashes at the delay time you want. Take the output of this and feed it into the BBD, take the output of the BBD and boost it so it will flash a second LED. Then it's the same deal as above, if the the delay time is what you think it should be the LEDs will flash in sync. Not very accurate, but it's good enough to tell you whether the chip is what it says it is.

If you want a readout of the delay time you could do this with a PIC or AVR chip, this could generate the clock and the signal for the BBD and measure and display the delay time on an LCD.

[armdnrdy]

Exactly!! Now we're on the right track!!

Maybe a rotary switch to change the timer rate for the different stage BBDs

[R O Tiree]

If all you're trying to do is tell whether a 3205 is really a 3205 and not a re-badged 3208 or similar scams, then how about using a couple of 74HCT4040 counters, drive the BBD chip with the required opposite-phase clock signals and use one of those clock signals to drive the counters. The counters can be made to count whatever number you want but, in this case, it's dead easy - for a 3205, you need to count to 4096. The first 4040 counts to 1024 and the second one counts to 4 = 4096. Press a switch to insert a voltage pulse (say 3V) into the delay chip input and that would also start the counter. Have an AND gate at the end and if 3V appears out of the end of the BBD at the same time as the counter has got to 4096, that drives a latch connected to an LED. You could have several LEDs to show 512, 1024, 2048 and 4096... A reset button to clear everything and you're good to go. You'd want several sockets to cater for the various types of chips (SAD1024, MN3205, MN3208, etc, etc) and make sure the power supplies are right... ZIF sockets so you don't bend any pins...

[armdnrdy]

R O Tiree,

Thanks for the reply!

Brilliant! Looking at the data sheet for the 74HCT4040 the timing diagram has the same stage designations as BBD ICs. This looks very promising!

I ask questions because....What I don't know DOES hurt me!! I was looking to go in a completely different direction.