Rectify sine wave signal

[armdnrdy]

I'm having a difficult time finding the proper way to rectify a single sine wave signal produced by a wein bridge oscillator. With a full wave bridge rectifier, the AC connects to two opposite corners of the bridge, the ground connects to the anodes between the AC connection, and the output is the two other diode cathodes.

The problem is that the AC signal produced by the wein bridge shares a common ground with the DC voltage, so there's no two AC inputs and ground.

The wein bridge that I'm using is a single supply design which utilizes a voltage reference. I was thinking since the sine wave amplitude travels between + volts and ground, that would make VR equal to 0 volts in this system. Can I then incorporate VR into the rectifier?

Does anyone know a quick and dirty way to change this signal to DC?

[amptramp]

Some stuff here:

http://www.niu.edu/~mfortner/labelec/lect/p575_01b.pdf

more here:

http://sound.westhost.com/appnotes/an001.htm

One idea they did not cover was the concept of using a comparator to determine whether the signal was above or below a reference and use it to switch from the input to the output of an inverting unity-gain amplifier via CMOS switches.  You need to determine whether you want halfwave or fullwave rectification (fullwave is more precise and requires less filtering but is more complicated).  These designs are precision rectifiers and diode clamps that use feedback to virtually eliminate the forward voltage of the diode(s).

[oldschoolanalog]

What is it you want to use the rectified AC for?
BTW, Boscorelli's "Stompbox Cookbook" has a real good chapter on V control.

[armdnrdy]

Hey Dave,

Thanks for the reply. I've been working on a BBD tester. The tester will show two things, that the BBD passes a signal and will show the number of stages. It involves switching different processes with 4053s among other things.

There was some thought that a BBD would pass a DC signal (delayed) which would turn off a counter/LED display driver IC. (CD40110) It looked promising when I bread boarded the BBD/clock portion of the circuit and scoped the output, but after reading through more BBD data sheets, I found that the usual cap found after the BBD output is there to filter the clock signal.

Having the square wave clock signal hitting the NPN switch would not work. Also, I'm not very sold on the DC bias voltage exiting the BBD delayed.

I went back to the original idea of inserting a sine wave from a wein bridge oscillator into the BBD. I need to rectify the signal when it exits the BBD so that it will saturate a NPN switch. The sine wave will make the transistor turn on and off.

[armdnrdy]

Thanks for the reply Ron,

I've looked over a few of the things that you linked already. I'm not sure if there's an easier way to rectify a single sine wave. I'm a bit leery of over engineering it! This signal doesn't have to be "pure as the driven snow", This isn't being used for an audio application. I just need it to stay away from 0 volts.

[PRR]

Why full-wave? A half-wave can have in and out common, is simple, and may tell you what you want to know.

[oldschoolanalog]

If it's MN3005 vs. MN3008 you are looking to test, any single BBD MN300X based analog delay and your ears will do the trick. With delay maxed there is a very audible difference between the two.
MN3007 vs. MN3009 can be tested in any MN300X based flanger. Again, a very audible difference. I have used these techniques to test many BBD's.
Can you please be a bit more specific as to what flavor of BBD's you are looking to test?

[armdnrdy]

Hi Paul,

Yeah, I was starting to lean towards half wave. This is kind of an unusual situation where the AC doesn't really have a dedicated common. What complicates the issue is that the wien bridge oscillator is a single supply design. I'm dealing with +V, VR, and GND instead of +V, 0V, and -V.

[armdnrdy]

Hey Dave,

The answer to your question is: All Flavors

I'm working on two models. The first one is for the positive supply MN32XX and the SAD1024. It will have two 14 pin, and one 16 pin ZIF sockets capable of testing 3205, 3208, 3204, 3207, 3209 and SAD1024.

The MN's and the test bed will be powered with +9 volts. The SAD will be powered with +12 volt directly. The data sheet states that the SAD can be operated at clock voltages down to 5 volts with reduced input bias and signal amplitude. I've made provisions in the circuit for that.

The second testor is for the higher voltage models. MN30XX and TDA1022. Once again three ZIF sockets on top of the enclosure that will test 3005, 08, 04, 07, 09, 10.

I was thinking to combine the two but it would involve too much switching and make the off board wiring a lot more complicated.

Now why am I designing this when I can pop open a stompbox, throw in a BBD and be done with it?

Well, for a few reasons. There are quite a few new builders out there that will not have a delay/flanger until they build it. Also, how many delays out there have relabled 3008's instead of 3005's and the builder doesn't even know it because he doesn't have anything to reference it to.

This tester as I mentioned before, will let you know if the BBD passes a delayed signal and if it is "real".

I usually source the harder to find parts before I commit to a build. Even though I've built quite a few delays and flangers, there are some BBDs that I have stocked for upcoming builds that I have no way to test. If any of the BBD's are "bad" it would be a lot more difficult to return them if they were purchased six months previously.

Some time ago, I built an ADA flanger based off of your and Charlie's work. (designed and routed my own board to fit in a 125BB! Sounds great) I sourced two SADs ahead of time. When I finished the build, I plugged one of the BBDs in.....nothing! plugged the next one in.....nothing! I sourced a third one from a different seller, plugged it in....guess what....nothing! I spent a week troubleshooting that thing! I checked everything! What could be keeping that signal from exiting the BBD?? It can't be all three BBD's

When the fourth one arrived I plugged it in and guess what....it worked! The first three were DOA! When I contacted the seller that I bought the first two from, he admitted that the cashe he found contained quite a few "bad" chips as he replace my two with the tested variety.

That's my hows and whys!

I guess when you've been bitten by a dog, you tend to carry a stick. This is my stick!

[Kesh]

you can use op amps

http://sound.westhost.com/appnotes/an001.htm

[oldschoolanalog]

@ Larry: Excellent points. I applaud your efforts and look forward to the results. 

[armdnrdy]

you can use op amps

http://sound.westhost.com/appnotes/an001.htm

Hey Kesh,

Thanks for the response. I decided to go with a 3 component half wave rectifier. (Schottky diode, load resistor, smoothing cap) As I stated earlier, this signal isn't being used for audio purposes. The sine wave signal has to pass through the BBD, through the rectifier, and hit the base of a switching transistor which will stop and hold a LED display counter/driver.

[armdnrdy]

@ Larry: Excellent points. I applaud your efforts and look forward to the results. 

Hey Dave,

I left a few things out of my hows and whys.

My first flanger build was a MXR 117 MN3007 retrofit clone. My first attempt at powering it up resulted in the BBD heating up to the point to where you could not touch it! Many hours of troubleshooting and MN3007 sourced from a different seller resulted in working flanger. (Bad BBD)

My first delay build was an AD-80. When I powered it up and adjusted the trimpots it worked! I was thinking, why do people hold this delay in such high regard? It doesn't seem to have that long of a delay! Luckily I had a Carbon Copy for reference. I set the delay of the Carbon Copy at the half way point, plugged in the two delays side by side to AB them. I was a bit surprised to find that the AD-80 delay wasn't even close to the 300MS it was supposed to be! (MN3008 relabeled to MN3005)

So you see with what I wrote before and what I just offered up.....unfortunately, I speak from experience!

Since I've been building stompboxes (which hasn't been that long) I've seen the BBD market drying up at a feverous pace. With the Chinese relabeling scams and with most sellers not testing the ICs before they sell them, I see a need for a BBD tester of this sort.