1K-1MHz Sinewave Generator Using one Freq. Pot

Get help with your Nuts & Volts projects or Kits purchased from Nuts & Volts here.
Post Reply
wizguru
Posts: 8
Joined: Mon Nov 23, 2009 3:20 pm
Contact:

1K-1MHz Sinewave Generator Using one Freq. Pot

Post by wizguru » Sun Dec 06, 2009 8:46 pm

I am trying to design a sinewave generator, needed as part of the electronics in 100 devices I am to provide a church. The sinewave generator must be cheap, use few parts (12 or less preferred) to be compact, be tunable from 1 KHz to at least 1 MHz using only one single pot, and use commonly available and inexpensive ICs and transistors. The sinewave output must be clean and with low distortion. My own experiments have produced some great squarewave, triangular wave and trapezoidal wave circuits which meet all of my other qualifications, but no sinewave generator. I have a ton of XR8038 and XR2206 ICs, but these are too rare and valuable for this application. From my research, the only low-parts decent sinewave generators I can find either use expensive parts or require ganged pots.

Please help.

Thanks, wizguru

User avatar
CeaSaR
Posts: 1626
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: 1K-1MHz Sinewave Generator Using one Freq. Pot

Post by CeaSaR » Mon Dec 07, 2009 9:57 am

I came across a similar question on the Instructibles site. What I found for that author was a link
to Velleman Kit's function generator. See this PDF.

Hope that helps you.

CeaSaR
Hey, what do I know?

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: 1K-1MHz Sinewave Generator Using one Freq. Pot

Post by Robert Reed » Mon Dec 07, 2009 12:18 pm

I don't think that will help you as it is not tunable and is only set for one frequency, probably fairly low. Any attempt to make it tunable would require multiple tuning elements and be an absolute nightmare to accomplish. Also amplitude and harmonic distortion would vary and probably be quite high. You made no mention of amplitude, and output impedance requirements. Just getting a pure sine wave alone is not an easy task, but to cover a 1000/1 to range is a real project. :sad:

dyarker
Posts: 1650
Joined: Fri Aug 22, 2003 1:01 am
Location: (somewhere), Afghanistan; from Rochester, NY
Contact:

Re: 1K-1MHz Sinewave Generator Using one Freq. Pot

Post by dyarker » Mon Dec 07, 2009 7:46 pm

1:1000 range in single sweep?

Perhaps some idea of what 100 copies of such a thing would be used for would bring up useful suggestions.
Dale Y

wizguru
Posts: 8
Joined: Mon Nov 23, 2009 3:20 pm
Contact:

Re: 1K-1MHz Sinewave Generator Using one Freq. Pot

Post by wizguru » Tue Dec 08, 2009 9:02 pm

The sinewave generators are to be used in church science class, where each young student has two of these modules, to teach these youngsters how to use various electronic test equipment, what a "pure" sinewave looks like on a scope, how sinewaves are involved in various electronic designs, and for audio, ultrasonic and music experiments. This is part of several other electronic projects I am doing for them, but the others have not stumped me like this one has. Amplitude must be decent enough to accomplish these projects, but need not be rail-to-rail, and the circuits will likely be used with various loads.

wizguru

User avatar
CeaSaR
Posts: 1626
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: 1K-1MHz Sinewave Generator Using one Freq. Pot

Post by CeaSaR » Tue Dec 08, 2009 9:32 pm

A bit more looking around and found a few more sites.

TalkingElectronics.com. Scroll down slightly on the left and click the 200 transistor circuits.
Book 1 pg 23-24 has several oscillators you could use.

Tutorial site. This page is about sine wave generators. You can use one of them.
Also, the main page of this site shows all the tutorials available.

CeaSaR
Hey, what do I know?

dyarker
Posts: 1650
Joined: Fri Aug 22, 2003 1:01 am
Location: (somewhere), Afghanistan; from Rochester, NY
Contact:

Re: 1K-1MHz Sinewave Generator Using one Freq. Pot

Post by dyarker » Wed Dec 09, 2009 8:01 am

Audio/music are generally considered to go up to 20KHz or 22KHz. Many ultrasonic remote controls operate a 44KHz. If you change your high end requirement from 1MHz to 80Khz or 100KHz, I think your chance of success will be much better, while still meeting your educational goal.
Dale Y

wizguru
Posts: 8
Joined: Mon Nov 23, 2009 3:20 pm
Contact:

Re: 1K-1MHz Sinewave Generator Using one Freq. Pot

Post by wizguru » Sat Dec 12, 2009 9:05 pm

Thanks for the links. However, no link provided describes a circuit which uses only ONE pot control to control the sinewave oscillator frequency, and that is a must requirement.

1 KHz - 1 MHz is important because they want an oscillator which can be expanded to possible other future experiments, such as heterodyning experiments. However, if a 1 KHz - 100 KHz circuit is possible which meets all of the other specifications, including one-pot control for tuning frequency and low distortion, let's see it. It will do for now to meet the educational requirements. The XR2206 is tunable from 0.01 Hz to 1 MHz, so 1 KHz to 1 MHz should be achievable.

Someone must know how to do this!

Thanks. wizguru

dyarker
Posts: 1650
Joined: Fri Aug 22, 2003 1:01 am
Location: (somewhere), Afghanistan; from Rochester, NY
Contact:

Re: 1K-1MHz Sinewave Generator Using one Freq. Pot

Post by dyarker » Sat Dec 12, 2009 11:42 pm

The XR2206 can generate 0.01Hz to 1MHz which is a range of 100,000,000:1. The XR2206 has a single sweep range of 2000:1. To achieve this the XR2206 has a LOT of transistors inside, all the components inside are on a single die for good thermal tracking amongst the components, some of the resistors inside could be laser trimmed during manufacture so they match. The XR2206 would use multiple transistors in the "current switches" section, so that one potentiometer on the outside could be the equivilant of multiple potentiometers on the inside.

To do what you want with discrete components and inexpensive op amps could easily end up costing 3 or 4 times the price of buying more XR2206s. (And wouldn't work as well either.) Basically, your design contraints are impossible. None of the links provided by others has a one pot solution because there is no such thing.

If you're sure you don't want to use XR2206s, and 1KHz to 100KHz (100:1) is enough, then I suggest an RC phase shift oscillator may (repeat may) be able to do it in one sweep. Maybe two sweeps with switched capacitors. Of course that requires 3 ganged pots. (Or 3 separate pots adjusted together manually.)

A sponsored link on google had XR2206s for $6.98. http://www.mouser.com has them for $3.08 in quantity 100. I would design both ways, add up the prices for the parts lists, then decide.

Cheers,
Dale Y

wizguru
Posts: 8
Joined: Mon Nov 23, 2009 3:20 pm
Contact:

Re: 1K-1MHz Sinewave Generator Using one Freq. Pot

Post by wizguru » Fri Dec 18, 2009 1:52 pm

Hello.

This is a donated project by me to the church. I cannot afford to pay $3+ a pop for XR2206 ICs to replenish my personal stock of them used for this project - $300+ total for 100. With all parts considered, this would bring up the project costs to an unaffordable $600+ ($6+ per circuit, including all other parts including a cheap box).

I just don't believe that no one is smart enough to design a paltry 100:1 1-pot sinewave generator of decent purity which will adjust from 1 KHz to 100 KHz using a few cheap opamps and transistors! I do not accept the premise that this is impossible! Where does it say that in the literature? Someone must be smart enough to have done this already. If we can put a man on the moon in the 50s but can't now design what should be a relatively simple and modest circuit, what does this say about us and how can we convince these kids into going into high-tech fields?

Thanks. John (wizguru (at) jjwill.com)

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: 1K-1MHz Sinewave Generator Using one Freq. Pot

Post by Robert Reed » Fri Dec 18, 2009 9:10 pm

wiz
The web is full of simple circuits and none of them work completely right. Once we start putting constraints on these circuits, complexity enters the picture. A Radio Shack oscillator is quite simple and isn't worth a hoot. An HP oscillator is a work of art and quite complex and will satisfy all needs. Like Rogers said ' There ain't no free lunches' :smile:

wizguru
Posts: 8
Joined: Mon Nov 23, 2009 3:20 pm
Contact:

Re: 1K-1MHz Sinewave Generator Using one Freq. Pot

Post by wizguru » Mon Dec 21, 2009 12:27 pm

Since I am unable to design this and no one else has presented a solution and I can't afford to pay for expensive parts these days, I will have to soon inform the church that I will have to un-volunteer for this project. Perhaps the kids can go into banking, sales, insurance, or become stock brokers or lawyers or something like that instead of growing up to become engineers and scientists.

Diodes, transistors and capacitors have non-linear characteristics which seems like they should be useful in producing sinewaves without using a complex and expensive circuit to do that. I tried a few combinations but could only get half of what looks like a good sinewave using a single pot. However, since I couldn't get the other half to produce a complete sinewave, I can't use my distortion analyzer to tell me how close I am.

wizguru

User avatar
CeaSaR
Posts: 1626
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: 1K-1MHz Sinewave Generator Using one Freq. Pot

Post by CeaSaR » Mon Dec 21, 2009 1:50 pm

Show us what you have come up with so far and it may be possible to get you part way toward our goal.
Perhaps as the kids get more excited, they may volunteer to help pay for expanding the capabilities of
the basic setup.

Is it a prerequisite that the sole control for the frequency is a pot? Would it be acceptable to use a switched
bank of capacitors to select a range? That would go long way toward your goal, although the cost ratio might
make it cheaper to use the actual higher end chips.

Tough one.

CeaSaR
Hey, what do I know?

dyarker
Posts: 1650
Joined: Fri Aug 22, 2003 1:01 am
Location: (somewhere), Afghanistan; from Rochester, NY
Contact:

Re: 1K-1MHz Sinewave Generator Using one Freq. Pot

Post by dyarker » Wed Dec 23, 2009 11:59 pm

Perhaps the kids can go into banking, sales, insurance, or become stock brokers or lawyers or something like that instead of growing up to become engineers and scientists.
The reverse of your position I would think! There are challanging things to be done in science and engineering. If it was too easy it wouldn't be exciting, then the brightest WOULD go looking for something else to do.

===================================

A low distortion sinewave oscillator isn't that easy, especially for variable frequency. The phase shift around the amplification/feedback loop must be zero degrees, 360 degrees, or multiple of 360 degrees, at the desired frequency. Also, the gain around the amplification/feedback loop must be one. If less than one, the signal gets smaller each time around till it stops all together. If greater than one, the signal gets bigger each time around till cutoff and/or saturation is reached; the signal becomes a sloppy looking square wave.

The amplifier (transistor or op amp) typically provide 180 degrees (inversion) of phase shift, leaving the other 180 degrees for non-active components. ((At RF frequencies variable caps and inductors are reasonable size and cost for 0/360 degree amp section (w/tank circuits))) A single pot combined with fixed resistors, caps and/or inductors simply can not vary the frequency very much without also changing the gain.

We're not keeping secrets from you, nor are we dumb; it is just physics and math. To recreate a small part of the functionality of the XR chip with cheap parts would take so many cheap parts the cost ends up greater.

I am sorry for the kids that this is out of budget. Your heart is in the right place. But what you want is impossible within the limits set.

Merry Christmas,

Dale
Dale Y

Dimbulb
Posts: 324
Joined: Sun Oct 13, 2002 1:01 am

Re: 1K-1MHz Sinewave Generator Using one Freq. Pot

Post by Dimbulb » Sun Dec 19, 2010 10:23 pm

In an analog sense the RC function for frequency has range limits because unlike resistors capacitors have a volume charecteristic.
So for freq 1k to 1M we need various sizes of uFarads. And So the best I can tell you is we dont have a smart capacitors.
pic vs 8038 is a matter of adaptation to what is available. hang on to those 8038s they're getting rare.

Post Reply

Who is online

Users browsing this forum: No registered users and 3 guests