Phaser phasing unevenly

[theonefromthesky]

Greets,

I'm not new to electronics, know my way around a board. But I am no engineer. Recently started making my own PCBS and circuits and all have worked straight out the box no problems.

http://www.tonepad.com/getFileInfo.asp?id=99

(Original link wrong, this link updated)

However, my Phaser is working but phases too quickly on high freqencies and too long on low freqencies, like an uneven sine wave.

I have made a few substitutions I used LM13700's instead of lm13600 and 0.0022uF instead of 0.0033uF caps.

Apart from that the circuit is the same.

Just wondering if the LM13700 and LM13600 are such vastly different packages that woudl be causing the issue?

Of course I have made sure of shorts, dry joints other obvious signs of damage etc but can't find any. I can't see any components that I have placed incorrectly, but shall double check shortly.

Any advice woudl be greatly apreciated!

Thanks

Jody

[Govmnt_Lacky]

It appears that you linked the Phase 90 project. That project uses JFETs to get phasing. You may want to repost the project link. Is it the Ross Phaser?

[theonefromthesky]

Bugger.... Your quite correct. yes it is the ross phaser!

http://www.tonepad.com/getFileInfo.asp?id=99

[Govmnt_Lacky]

Cool... now that the link is fixed, I suggest that you read through the following link:

http://www.diystompboxes.com/smfforum/index.php?topic=29816.0

Take 5 minutes and verify ALL component values, placement, and verify you installed ALL JUMPERS!!!

After that, you should repost with IC voltages, etc. Maybe even some pictures.

You should get it corrected quickly after that.


Good Luck  

[Mark Hammer]

I've made a bunch of these, and I have to say that some sweep very nicely and some don't.  The key, I think, lies in the 270k/100k pair that provides the bias voltage on pin 3 in the LFO, as well as the value of the 270k resistor on pin 1.  The pin 1 resistance partly determines the shape of the sweep, but so does the bias voltage.  I keep tinkering with those on units where I'm not happy with the sweep, but I've never stumbled onto something that is 100% dependable.

So, no, you're not nuts or inept.  Do measure the precise resistance of those components for us, as well as the specific bias voltage at pin 3, using those components, and let us know what they read.  That might provide some useful clues to being able to get this thing right ALL the time.

[theonefromthesky]

Thanks, for the input guys. I was having a read through pther phaser problems, and I can see that people are having similar issues.

I would have thought that the natural varience in the majority of components is what is pushing the sweep off!

I shall measure some voltages, but I have to reasemble first, as I have taken it out of the box to paint, thus not working.

Also, on a side note. I saw that it needed Rev Log pots, to which I do not have any and currently using 500kB Lin type. Can I modify to the response of Lin pots to create log as I can't locate Rev log. I know I could just use somethign like a 100k lin to emulate the last part of the curve, but it would cut off the start of the long sweep time which I like!

Shall get back momentarily.....

Jody

[Mark Hammer]

A reverse-log pot is simply a log pot but in the opposite direction.  If you don't mind having a control that goes slower as you go clockwise, just wire up a normal log pot.

Alternatively, install a SPST toggle to throw a 51k-68k fixed resistor across a 500k lin pot for fast/slow ranges.  The need for the anti-log pot is basically about being able to dial in the faster speeds with some accuracy and repeatability.  The under-50k range is where the faster speeds live, so if you can switch between a 500k and 50k-ish pot, that will allow dialability of both fast and slow.

[johngreene]

I think you will find that 180K and 100K work better than 270K/100K on pin 3 for the LM13700. It's not only providing a bias voltage but because of the feedback that comes from Pin 8 and 5, it provides the necessary hysteresis for the oscillator to work without hitting the rails. You should be able to get it to work as needed by only adjusting the 270K resistor connected to pin 3. At least this has been my experience with the LM13700. 
Also, because you are using a smaller capacitor, you may have to change the value of the 10k that is in series with the LFO output to compensate. The OTA is basically acting as a variable resistor that works against the capacitor (ala univibe) and its 'inverted' resistance is dependent on the current being injected into pin 1/16. So smaller capacitor means the center resistance point is higher which means the center point of the LFO current input needs to adjust to compensate.
I hope that makes sense.

[theonefromthesky]

yes it does, thanks to you both!

Last night I had a little play with the circuit and put a pot instead of both resistors. I adjusted until it sounded even.

Low and behold the values my pots were at are 105k and 156k, so a 100k and 150k resistor later. It sounds dandy! Its also created an interesting effect when the rate is turned right up!

I must thank you for the ideas, as I would never have figured it out on my own. My knowledge of electronics is but very limited. But now armed with further knowledge, I can at least have some idea where to start if a similar problem comes up in the future!

Thanks again!

Jody