New Looper Design

[The Tone God]

yes it's true and now is my problem ! I've built a looper with isd2564 with slacker design control and when I switch from record to play the chip cuts some notes at the begin of the sample , what's wrong???

When we use the IC just as a plain recorder the time it takes the IC to come out of power down, latch the mode/address and P/R inputs, set the address pointer, configure the needed modules, and start the function should still not take that long so I don't think this is the fault of the IC.

I assume you are using the inverter based digital circuit. Just looking at that design I think the delay is coming from the large debouncing network on the gate inputs for the PD pin. You could try to tweak the values to speed it up but you could get issues if you set it to fast.

I have finished the preliminary digital logic design for my looper which I designed from scratch. Here some of the highlights.

• It will only need two DPDT switches. Bypass and record.
• True bypass.
• Puts the IC into sleep mode when bypassed.
• Starts the loop from the beginning when engaged from bypass.
• You can reset the loop by cycling the bypass.
• It can be configure to record in bypass but you lose the "true bypass" feature. The input impedance is pretty high so it should not be an issue.
• If you hit record during playback it will stop the loop playback and start recording the new loop.
• If playback is engaged when coming out of record the new loop will start almost immediately.
• Indicator LED that blinks on at the end of each loop when in playback. The indicator will go solid when record is engaged.
• Safe Switch to disable record function.
• The logic is not record IC family specific so it should work on most of the record IC families including the 1xxx and 25xx series.
• Requires only a single logic gate IC.
• A few other interesting features not yet mentioned.

Yes a non-optocoupler based vibrato is on its way (I hope) and the output pop issue seems to be worked out. As of this time the audio section will probably be fairly simple too.

Enough about this. I don't want to hijack the thread and I have to get back to work on it anyways.

Andrew

[slacker]

yes it's true and now is my problem ! I've built a looper with isd2564 with slacker design control and when I switch from record to play the chip cuts some notes at the begin of the sample , what's wrong???

When we use the IC just as a plain recorder the time it takes the IC to come out of power down, latch the mode/address and P/R inputs, set the address pointer, configure the needed modules, and start the function should still not take that long so I don't think this is the fault of the IC.

I assume you are using the inverter based digital circuit. Just looking at that design I think the delay is coming from the large debouncing network on the gate inputs for the PD pin. You could try to tweak the values to speed it up but you could get issues if you set it to fast.

I'd agree with that. According to the datasheet the ISD2564 only takes 50 milli seconds to power up, this is longer than the ISD2560 that I tested with but it's not long enough to cause any problems. How much time are you losing at the start of the sample? My switch probably takes a few 10ths of a second to switch so if you're losing more than that your problem is not with the switching. You can try making the 2 2.2uF caps smaller, this will speed up the time taken to switch.
While doing the samples I posted I did notice it was quite tricky to get a nice smooth loop going, but I couldn't decide whether it was a technical issue or my bad timing I'll take another look at the design and see if I can improve it.
Or you could wait for The Tone God's looper thats bound to be good

[domenico]

hi thanks for the replys , I've tried 1mf cap with no results...should I replace 2564 with 2560 ?
thanks
domenico

[The Tone God]

Or you could wait for The Tone God's looper thats bound to be good

Thanks for the vote of confidence.

hi thanks for the replys , I've tried 1mf cap with no results...should I replace 2564 with 2560 ?

I am using a 2575 which I have little if any trouble with start delay and the difference in times between the 2564 and 2675 is roughly 10-15ms which I think is probably not enough to make a difference in performance.

One thing I will make note of the 2564 is it has a low cut off frequency (1.7kHz) and low sampling rate (4.0 kHz). If you want to get performance similar to the 1420 (6.4kHz sampling and 2.6kHz cutoff) I would go with either the 2540 (6.4kHz sampling and 2.7kHz cutoff) or the 2575 (6.4kHz sampling and 2.7kHz cutoff).

While all that extra time sounds like a waste I am hoping to give my looper the ability of multiple loops so the extra time won't be wasted. Oh and I got the vibrato working.

Andrew

[zachary vex]

october '05 till now... sounds like my development cycle for the lo-fi loop junky!  8^)  actually, it was a couple of years because i kept putting it aside and trying new things after a few months of letting it sit.  at one point i completely gave up on it, but in the end, i had several important breakthroughs including discovering how to eliminate popping at the output.  or most of it, at least.  8^)

[The Tone God]

Roughly one dozen DIY members + nearly a year = 1 Zachary Vex



Andrew

[zachary vex]

your challenge, of course, will be to fit it on a circuit board .88" wide by 3.6" long.  8^)

[The Tone God]

your challenge, of course, will be to fit it on a circuit board .88" wide by 3.6" long.  8^)

Hmm...well if I don't include every feature I am going to suggest in the article I think I can do it. I sir accept your challenge.

Andrew

[nelson]

Tiny caps, 5 layer board, resistors standing, 9mm pots: NO PROBLEM.

[The Tone God]

Tiny caps, 5 layer board, resistors standing, 9mm pots: NO PROBLEM.

5 layer board, nah. Not worth the time and expense of doing a PCB just for a one-off lark. I'll perf it. I'll have to do some CADing to see what pots I can fit in there but I figure I can fit five easily maybe six if I'm careful. I have a bunch of other dirty tricks that I can use as well.

Andrew

[domenico]

now my looper works ! It mounts a 2560 chip with slacker control and a bypass/reset switch .
the looper records and plays in time only when after a recording you reset the looper with bypass/reset
switch
to play the loop you have to set the looper in play and to hit the reset/BYPASS SWITCH
the problem of notes cutting remains when after the recording you set the looper on play without resetting it with  reset/bypass switch
bye!

[zachary vex]

i did it with only two sides, but i had to use heavy copper to eliminate noise issues.  keep that in mind.  8^)  hey, doesn't my lo-fi loop junky do everything on your list except go for longer than 20 seconds?  i guess the way i got it to fit on that small board was by not needing a logic chip... the family i used has enough on-board logic to use simple discrete component tricks to make it do what i wanted.  i always wanted a longer looper, though... good luck, tone god!  8^)

[The Tone God]

hey, doesn't my lo-fi loop junky do everything on your list except go for longer than 20 seconds?

On that list yes. I will be hopefully including a few other features as mods like a phase switch for the recorded output so if you play along you can get a different chorus effect. The big feature I hope to get working is the multiple loops as that is one the big reasons why I went with the 25xx series. I have a few other minor features I want to play with.

The controls on my current looper are also more extreme. The vibrato can for example slow down the playback to the point of almost stopping, but not reseting the IC, without causing too much added time to the loop. It sounds like a record starting up. The loop input and output volume controls also have more so the IC's input can be overdriven or the output could overdrive other effects/amps. There is also a final output control so one could mix the loop and dry signal to their tastes.

i guess the way i got it to fit on that small board was by not needing a logic chip... the family i used has enough on-board logic to use simple discrete component tricks to make it do what i wanted.

I took a quick look through the other family datasheets and I can see why it would be alot easier to use one of those. Those families are really optimized for pushbutton operation where the 25xx series is trying to cater to both pushbutton and uC needs. To do that they removed some of the internal logic and exposed more of the internal controller's inputs and outputs. For this reason I have generate some of the logic externally to get back the functionality. In the end it has only taken three gates, three resistors, and two caps to do the logic.

While I'm probably losing space with the external logic I'm probably getting some back in other areas. For example I'm using the last gate as the clock to drive the vibrato. Besides using a unused gate to save space I can add the feature that the vibrato will be automatically disabled during recording and enabled during playback. I also can do a power down / true bypass trick using only a DPDT switch with the addition of only one part.

The other big problem is that if I'm going to do multiple loops I need access to the address pins which mean I can't use any of the internal operational modes. Hopefully I can get the loops to work so that as you are in the middle of one loop's playback you can select another loop and when the current loop reaches the end, or you cycle the bypass switch, it will jump to the selected loop allowing different phrases to be stored.

The logic took me a couple of days to get working right. I could have grabbed a uC and done it in minutes but that would be cheating. If I had decided to go with a uC I would have grabbed one the other IC families and done some really cool stuff. The output portion ALOT easier too.

i always wanted a longer looper, though... good luck, tone god!  8^)

Thanks Mr. Vex. I haven't done a small build awhile so this should be fun.

Anyways back to work.

Andrew

[robbiemcm]

Good to see this topic has come back to life.

Tone God: yours seems really interesting, and you certainly have a lot of really good features lined up for it.. all of which I can see as being useful. Please don't sacrifice these features in order to cut down on the size though, I know I personally would rather something can do even just a little more a few extra centimeter of floor space. I intend to build yours if I survive Jaicen's looper first

[The Tone God]

Tone God: yours seems really interesting, and you certainly have a lot of really good features lined up for it.. all of which I can see as being useful. Please don't sacrifice these features in order to cut down on the size though, I know I personally would rather something can do even just a little more a few extra centimeter of floor space. I intend to build yours if I survive Jaicen's looper first

I guess I'll give a little bit of an update. The circuit is finished, tweaked, and works as expected. I have a rough schematic drawn and the article is partially done. I got all the extra features working including the multiple loops. I am content with it.

I'm not going to be offering a PCB so the size thing is not an issue. I made the decision awhile back that for now on I will only give the basic schematic with a functional explanation and some mod suggestions. From there people can decided what they want to do.

The size challenge is only a silly little thing on the side that I am doing for fun. I would not inflict those restrictions on other builders. As for the challenge I am not concerned about getting the basic version in there. The only question is how much more can I fit in there which the answer according to my design is just alittle bit more. This would not be a pedal going on my "board" anyways so I don't care if I don't get every feature in there. I already have the design completed and I actually just got the parts I needed to start building mine. I'll post pics when I post the article.

Andrew

[robbiemcm]

Sounds good. Although I must say I'm not looking forward to making my own PCB for it I've only ever done PCBs for two things, Easy Drive and Obsidian Overdrive. Either way, I can't wait to have a looper to play around with. Make sure you make a link to a new topic if you start one when the project is done as I don't check the rest of the forums much these days.

[The Tone God]

Sounds good. Although I must say I'm not looking forward to making my own PCB for it I've only ever done PCBs for two things, Easy Drive and Obsidian Overdrive. Either way, I can't wait to have a looper to play around with.

I'm perfing mine. Maybe someone else will make a board as there are already a few that can be used as a base.

Make sure you make a link to a new topic if you start one when the project is done as I don't check the rest of the forums much these days.

Will do. I'm right now in the middle of writing the article.

Andrew

[nelson]

This is one that I will be doing a PCB for.

If we get our resident Vero boarders to help this could be very accessible.

Not having seen the schematic, I am hoping I can fit this in a 125 style enclosure with 9mm PCB mount pots.

The board mounting vertically in the enclosure as opposed to horizontal. Leaving space for 2 footswitches, jacks and battery.

[ethrbunny]

Gah! This is killing me. I have two 2560s waiting for a schem. Id pay for a PCB! Heck. Im about to pay for a commercial looper  ...

[Michael Allen]

*shakes off the dust* Ahh.... I might have to give this a shot. Havn't made a layout in quite some time.  Actually the Vanishing Point was the last project I even worked on, what, a year and a half ago? Back in the saddle... This looper is gonna be tight!