Echo Base - a new PT2399 delay

Started by slacker, August 27, 2007, 04:33:19 PM

Previous topic - Next topic

slacker


gdk87tyaw67w

Hi, long time reader, first time poster here. I read through the topic, including the discussion about integrating a tap tempo module, which is of particular interest to me. However, I understand that the Echo Base's modulation section doesn't get along well with tap tempo, and that a non-DC coupled one (like from the Magnus Modulus) would. Is that correct? Should I expect any other problems with it, or is it expected to work more or less seamlessly?

The reason I'm asking is I'm building a PT2399-based delay (the Rebote 2.5 circuit, as I already have a PCB for it), and I'm planning on adding tap tempo to it. It's my own design, though pretty similar to the PTAP (except using an LED/LDR combo instead of a digital potentiometer). After reading this topic, though, I'd like to add a modulation section to my delay as well, so now I'm trying to figure out the best way to do that.

Thanks in advance for any advice.

Taylor

Do you think an LED/LDR will be accurate enough to work for this application? Seems like you need precision that they don't have.

slacker

#703
Welcome to the forum :)

The issue with the PTAP and the Echo Base is that the PTAP is calibrated assuming a resistance of 1k between pin 6 of the PT2399 and the digital pot. The transistor used for the modulation on the Echo Base basically throws the calibration out.
The other issue was the change in DC level on the transistor caused by the modulation, but this can be solved by added a capacitor between the output of the LFO and the mod level pot.
For an individual build I would think you could just calibrate your set up to take into account the resistance of the transistor and use the Echo Base method.

gdk87tyaw67w

Thanks for the info, slacker!
I'm not sure I understand the calibration issue completely. The resistance is what determines the delay time, right? The transistor basically varies that resistance periodically, which provides the modulation. How does one calibrate for variable resistance?

Quote from: Taylor on September 18, 2009, 02:19:58 PMDo you think an LED/LDR will be accurate enough to work for this application? Seems like you need precision that they don't have.
Unless I'm mistaken, an LED/LDR will have virtually infinite precision, on account of being an analog device. A digital pot, on the other hand, has a finite set of values. It, too, is sufficient for this particular application - given a maximum delay time of 500ms, a digipot with 8-bit output has a precision of 2ms.
For what it's worth, I'm planning on driving the LED/LDR from the microcontroller's PWM output.

slacker

#705
Quote from: Indrek on September 18, 2009, 08:29:41 PM
I'm not sure I understand the calibration issue completely. The resistance is what determines the delay time, right?

As I understand it, the PTAP is programmed so that it gives you the correct resistance for a particular delay time taking into account the 1k of resistance in series with the digital pot. If you change the value of the 1k resistor then you wont get the correct delay time. On the Echo Base the resistance of the transistor when the modulation is off will not be 1k so the PTAP won't work properly. I haven't tried it, but if the resistance of the transistor is less than 1k you could just put a resistor in series with it to make the total 1k, then the PTAP would work when the modulation was off. If the resistance is more than 1k you could adjust that as well by playing with the 39k resistor in parallel with the transistor, but that might mess up the modulation effect.
For your design you could just measure the resistance of the transistor with the modulation off and take that into account when programming your system. Or have some way of adjusting the total resistance to take into account variances from pedal to pedal, you'll probably have to do that anyway if you're using a LED/LDR combo.

Quote
How does one calibrate for variable resistance?

I don't think you need to really. If you're only adding small amounts of modulation then the timing should still be accurate enough. If you're adding larger amounts of modulation then you're not really likely to be bothered about any sort of accuracy, you probably just want the tap tempo to give a very rough delay time.
To be honest personally I can't see the point of tap tempo when using modulation.

Quote
Quote from: Taylor on September 18, 2009, 02:19:58 PMDo you think an LED/LDR will be accurate enough to work for this application? Seems like you need precision that they don't have.
Unless I'm mistaken, an LED/LDR will have virtually infinite precision, .

A LED/LDR is analogue but they might not be predictable enough in how they change resistance for you to be able to program the chip with any accuracy. Even if you can do that, the PWM output is digital anyway so you will only have what ever precision you can get from that.

gdk87tyaw67w

#706
Quote from: slacker on September 19, 2009, 07:01:40 AMI don't think you need to really. If you're only adding small amounts of modulation then the timing should still be accurate enough. If you're adding larger amounts of modulation then you're not really likely to be bothered about any sort of accuracy, you probably just want the tap tempo to give a very rough delay time.
My thoughts exactly. I'm planning on using only small amounts of modulation most of the time, plus even if the individual delays are a bit off, they're still centered around whatever the tempo is, so it shouldn't be a problem.
I think I'll even use trimpots for the mod depth and speed and preset them to fairly low values.

QuoteA LED/LDR is analogue but they might not be predictable enough in how they change resistance for you to be able to program the chip with any accuracy. Even if you can do that, the PWM output is digital anyway so you will only have what ever precision you can get from that.
True, though PWM outputs usually have at least as good resolution as digital pots, at least from what I've seen. Plus, I'm having a hard time finding a digipot with 50k resistance and decent resolution.
Anyone know the specs of the pot that PTAP uses?
Edit: found it. MCP41050. Unfortunately it isn't available where I live.

Valoosj

Ian, I have another situation for you.

After the first repeat, the next repeats become very quiet, no matter how I set the feedback and volume knob. Is there anything I can tweak to make the 2nd, 3rd, ... repeat the same volume as the 1st? My feedback resistor is now 20K, as I am using the extra time board. Maybe I should try increasing this one? Or wouldn't that solve the problem and just cause it to oscillate a lot sooner? (with the 100K the delay oscillates very easily)
Quote from: frequencycentral
You squeezed it into a 1590A - you insane fool!  :icon_mrgreen:
Quote from: Scruffie
Well this... this is just silly... this can't fit in a 1590B... can it? And you're not even using SMD you mad man!

gigimarga

Hello,
Anyone has a PCB for the true-bypassed version of this fabulous delay?

Thx!

gigimarga

Quote from: gigimarga on September 26, 2009, 01:56:16 PM
Hello,
Anyone has a PCB for the true-bypassed version of this fabulous delay?

Thx!
???

Valoosj

Quote from: Valoosj on September 25, 2009, 05:57:14 PM
Ian, I have another situation for you.

After the first repeat, the next repeats become very quiet, no matter how I set the feedback and volume knob. Is there anything I can tweak to make the 2nd, 3rd, ... repeat the same volume as the 1st? My feedback resistor is now 100K, as I am using the extra time board. Maybe I should try increasing this one? Or wouldn't that solve the problem and just cause it to oscillate a lot sooner? (with the 100K the delay oscillates very easily)

Made a mistake in my explanation. And I just replaced that 100K with a trimpot, but this resistor is not causing the issue that every repeat after the 1st one is too silent.
I really don't know what I can try now to fix this.
Quote from: frequencycentral
You squeezed it into a 1590A - you insane fool!  :icon_mrgreen:
Quote from: Scruffie
Well this... this is just silly... this can't fit in a 1590B... can it? And you're not even using SMD you mad man!

slacker

Quote from: Valoosj on September 25, 2009, 05:57:14 PM
After the first repeat, the next repeats become very quiet, no matter how I set the feedback and volume knob. Is there anything I can tweak to make the 2nd, 3rd, ... repeat the same volume as the 1st?

I guess this must be something to do with the extra time mod, because the original doesn't have this problem. I'll have a think about what the problem might be.

Quote from: gigimarga on September 26, 2009, 01:56:16 PM
Anyone has a PCB for the true-bypassed version of this fabulous delay?

You can build the true bypass version on the normal PCB, you just leave out the switching stuff. I could do a schematic showing which parts to remove if that will help.
You can also just build the normal version and leave out the boss/tails switch, that makes it "always on" then just wire it true bypass. 

gigimarga

Thx a lot slacker!
I've built it normal and I like it a lot, but I'm looking for a smaller PCB, that's what I want a PCB without switching :)

Valoosj

Gigimarga, on one of the first pages there was somebody that fitted it into a 1590B (zvex type).
Quote from: frequencycentral
You squeezed it into a 1590A - you insane fool!  :icon_mrgreen:
Quote from: Scruffie
Well this... this is just silly... this can't fit in a 1590B... can it? And you're not even using SMD you mad man!

gigimarga

Quote from: Valoosj on September 29, 2009, 05:01:48 PM
Gigimarga, on one of the first pages there was somebody that fitted it into a 1590B (zvex type).

Thx Valoosj!
I looked almost on all pages and I didn't find any layout...only a picture of a PCB from Auke Haarsma...I will look again when I will be more fresh :)

Valoosj

I think it was Chawk with the smaller layout, but his pics are gone.
Quote from: frequencycentral
You squeezed it into a 1590A - you insane fool!  :icon_mrgreen:
Quote from: Scruffie
Well this... this is just silly... this can't fit in a 1590B... can it? And you're not even using SMD you mad man!

gigimarga

Quote from: Valoosj on September 30, 2009, 04:54:25 AM
I think it was Chawk with the smaller layout, but his pics are gone.

Thx a lot Valoosj!
I will try to find one...no hurry :)

Valoosj

Quote from: slacker on September 29, 2009, 04:04:51 PM
Quote from: Valoosj on September 25, 2009, 05:57:14 PM
After the first repeat, the next repeats become very quiet, no matter how I set the feedback and volume knob. Is there anything I can tweak to make the 2nd, 3rd, ... repeat the same volume as the 1st?

I guess this must be something to do with the extra time mod, because the original doesn't have this problem. I'll have a think about what the problem might be.


Just wondering if you have had the time to have a look at this problem  :)
Quote from: frequencycentral
You squeezed it into a 1590A - you insane fool!  :icon_mrgreen:
Quote from: Scruffie
Well this... this is just silly... this can't fit in a 1590B... can it? And you're not even using SMD you mad man!

Nad

Hi,

Long time lurker, first time posting.

I wanted to say thanks to Slacker for sharing the Echo Base and also for all the debugging advice.   I've just built an Echo Base using the Vero layout and this thread has been invaluable.  

I love this pedal and the sounds that can be got from it.  

I now plan to build another Echo Base with true bypass

Thanks again.

FlyingZ

Sorry for my laziness in not reading the post. Is there a version without the mod section but with a nice tone control on the repeats?