A/DA Flanger retrofit with MN3007

[Rocket Roll]

Tahnks in advance! Looking forward to that!

[snap]

Thanks, Dave. I'm kinda tied up today, but should be able to post the new ADA layout, updated to use either 3007 or 3207, within the next couple of days.

~ Charlie

it`d be cool!

[Prive]

Is there any chance that this daughterboard could be adapted to allow the installation of the MN3007 into other SAD1024-based effects?

While this was not designed as a "one size fits all" retrofit; I'm confident some adaptation could be made to allow the MN3007 to replace the SAD1024 in other circuits. Anything specific? (EM/DEM? )

Dave

I have repaired my MXR flanger already but will be fun to try this retrofit, what d'you think? could it work in my MXR?

[Auke Haarsma]

Looking very very much forward to it Charlie. You seem to be Sinterklaas! (http://en.wikipedia.org/wiki/Sinterklaas)

[StephenGiles]

Ever seen a clock + buffer using LM339 + CMOS4069? This seems to use 3 gates in parallel to drive 2 x MN 3005.

http://www.film-tech.com/warehouse/manuals/KINTEKKT24.pdf

Have you checked this out?

[StephenGiles]

4047 into SAD 4096
http://www.film-tech.com/warehouse/manuals/SMARTSR330.pdf

......and this?

[moosapotamus]

Those links are interesting, Stephen. I particularly like the simplicity of the second one.

I have repaired my MXR flanger already but will be fun to try this retrofit, what d'you think? could it work in my MXR?

You won't know until you try. What have you got to loose?

~ Charlie

[oldschoolanalog]

Ever seen a clock + buffer using LM339 + CMOS4069? This seems to use 3 gates in parallel to drive 2 x MN 3005.
http://www.film-tech.com/warehouse/manuals/KINTEKKT24.pdf

Have you checked this out?

I was looking at this (pg. 19) and am curious to know why would one need six inverters in parallel for each clock phase?
Each clock phase has it's own 4069! And unless I'm seeing things (always a good possibility), aren't they configured 6 in/6 out?
Anybody have any thoughts? 
Stephen, how/where do you keep coming up with all this great stuff? 

I have repaired my MXR flanger already but will be fun to try this retrofit, what d'you think? could it work in my MXR?

I have a bunch of those old 117's kicking around. I'll take a look at the schem & pedal later. 

Dave

[StephenGiles]

Ever seen a clock + buffer using LM339 + CMOS4069? This seems to use 3 gates in parallel to drive 2 x MN 3005.
http://www.film-tech.com/warehouse/manuals/KINTEKKT24.pdf

Have you checked this out?

I was looking at this (pg. 19) and am curious to know why would one need six inverters in parallel for each clock phase?
Each clock phase has it's own 4069! And unless I'm seeing things (always a good possibility), aren't they configured 6 in/6 out?
Anybody have any thoughts? 
Stephen, how/where do you keep coming up with all this great stuff? 

I have repaired my MXR flanger already but will be fun to try this retrofit, what d'you think? could it work in my MXR?

I have a bunch of those old 117's kicking around. I'll take a look at the schem & pedal later. 

Dave

I always save links of sites I find so that I can go back to them. The Film-tech site used to list the additions but no longer does.

[moosapotamus]

Okay... finally got a first pass done on the new layout. Here it is...
http://moosapotamus.net/ADAflangerOVRmn3007.GIF

None of these boards have been made yet. But using a MN3007 has been tested with OSA's retrofit in one of the previous SAD1024 boards and it sounds great! Using one of the 3207 chips (BL3207, V3207) has not been tested yet, at least not to my knowledge. But this new layout is setup with pads that can be jumpered two different ways for either a 3007 or a 3207. There is also a spot for a +9V regulator to go with the 3207. If using the 3007, a jumper can be installed between the outside pins of the regulator.

I would really appreciate another set of eyes on this.

Here's a detail of the area in question...


Q2 is a 78L09 +9V regulator that can be installed if using a 3207 chip.
Does anyone think there should be any additional components for power filtering for the +9V regulator?

The jumper pads work like this...
Pad P is V+.
Pad R is ground.
Pad T is the leg of the trim pot.

If using the MN3207 chip jump P to P1, R to R1, and T to T1.
If using a 3207 chip, jump P to P2, R to R2, and T to T2.

Please let me know if that makes sense and/or if anything else, above, looks mixed up.

Thanks
~ Charlie

[Zben3129]

I think there may be an issue with using mn3207 in this circuit but I would need a look at the schamatic to know for sure, and can't find it right now  Could you link me to it?


As you know, mn3207 has a few differences from mn3007....3207 runs on + voltage supply, and can only run on 9v maximum. These differences have been covered and should not pose a problem.

However, the mn3207 needs a positive clock frequency (between 0v and 1.1v IIRC) while mn3007 needs a negative clock frequency (between 0 and -1.1v IIRC). This is the reason mn3101 is used to drive mn3007, while mn3102 is used to drive mn3207; mn3101 provides a positive clock frequency while mn3102 provides a negative clock frequency.

The BBD in this circuit, whether it be sad1024 or mn3007/3207 is driven by a CD4047 (or some CMOS chip, can't remember part number), correct? If so, it would also be necessary to alter the clock stage to make the frequency positive rather than negative. I am not sure if this is possible with the chip used as the clock driver, and if it is possible I am also not sure if it is practical.


As always, if someone sees anything wrong with this speak up so that the mn3207 issue can get sorted out correctly.


Zach

[RedHouse]

Does anyone think there should be any additional components for power filtering for the +9V regulator?

Moose, if you can fit the cap's in that would be good, I have experienced random issues/problems in the past with the 78xXX series when running w/o capacitors, not every time but just when you don't need it if you know what I mean. A couple .1's will do nicely and they can be those itty bitty little blue 50v ceramic mono's to save space.

In fact now that I mention it, one can really just piggy-back them on the solder side if there is no real estate for Wima style caps on the board.

-Brad

[snap]

SMD !

[moosapotamus]

I think there may be an issue with using mn3207 in this circuit but I would need a look at the schamatic to know for sure, and can't find it right now  Could you link me to it?

Thanks for taking a look at this Z. There is a link to the schematic in the first post on page 1 of this thread but, to make it easy,  I'll put it here too.

ADA clone schematic w/ MN3007

And there are schematic snippets showing the alternate connections for the MN3207 at the bottom of page 3 of this thread, here...
http://www.diystompboxes.com/smfforum/index.php?topic=72329.msg587202#msg587202

I know that the MN3007 will do the trick (thanks again OSA!). But I'm not really sure, myself, if the 3207 can work as shown. I'm relying on everyone's kindness to figure out if it makes sense to include the 3207 option. If not, it would simplify the layout to just have the 3007. Anyone else have any thoughts on this?

Moose, if you can fit the cap's in that would be good...

So, you don't think the single 22uF (or 33uF) will do it? I might be able to find some room for a couple of Wima 0.1uF caps to put in parallel with the 22uF.

Anything else?

Thanks
~ Charlie

[RedHouse]

I think there may be an issue with using mn3207 in this circuit but I would need a look at the schamatic to know for sure, and can't find it right now  Could you link me to it?

Thanks for taking a look at this Z. There is a link to the schematic in the first post on page 1 of this thread but, to make it easy,  I'll put it here too.

ADA clone schematic w/ MN3007

And there are schematic snippets showing the alternate connections for the MN3207 at the bottom of page 3 of this thread, here...
http://www.diystompboxes.com/smfforum/index.php?topic=72329.msg587202#msg587202

I know that the MN3007 will do the trick (thanks again OSA!). But I'm not really sure, myself, if the 3207 can work as shown. I'm relying on everyone's kindness to figure out if it makes sense to include the 3207 option. If not, it would simplify the layout to just have the 3007. Anyone else have any thoughts on this?

Moose, if you can fit the cap's in that would be good...

So, you don't think the single 22uF (or 33uF) will do it? I might be able to find some room for a couple of Wima 0.1uF caps to put in parallel with the 22uF.

Anything else?

Thanks
~ Charlie

The datasheet indicates a .33 on the input and a .1 on the output but it assumes a reasonable noise/ripple free supply.

I didn't see if you have the 78L09 being fed from the 7815 or from the main power input to the board, if it's from the 7815 it should only need the small caps .33/.1 if it's from the main power input the 22uF would be better for sure.

[oldschoolanalog]

A couple of notes.
This should read:
If using the MN3007 chip jump P to P1, R to R1, and T to T1.
If using a 3207 chip, jump P to P2, R to R2, and T to T2.
If anybody claims to have proofed this layout and didn't notice this: Shame on you!!! 

Also,(and I noticed this right before I clicked "post") the 1k R in the V divider is hardwired to ground as shown. For the 3207 setup this needs to be lifted from gnd so it can go to Vdd (pin 5, 3207). The 14k R to ground in the 3207 setup is correct. Connecting points R & R2 will result in both sides of the V divider being connected to ground. Might want to reroute some of them traces.
Other than that; great job. 

I have to agree that the 78L09 reg should have some provision for filter caps. Better safe than aggravated later. Maybe moving C33 and that trace under it "south" just a bit would free up enough room for a 1u at the input and a .1u at the output of the reg? Might be kind of tight. But doable.

I know that the MN3007 will do the trick (thanks again OSA!). But I'm not really sure, myself, if the 3207 can work as shown. I'm relying on everyone's kindness to figure out if it makes sense to include the 3207 option. If not, it would simplify the layout to just have the 3007. Anyone else have any thoughts on this?

I think it's a good idea to include the 3207 option as it would probably make alot of folks very happy. Plus, it's (almost) already done! On paper it should work. However my one concern; if it works; is how is it going to sound? I can say with confidence that the 3007 circuit has all the A/DA "sonic character" of the 1024 & 3010 versions I compared it to. I was going make a retrofit board for the 3207 but then realized I had none in my BBD stash. Maybe somebody else wants to try a 3207 retrofit board? It's really not all that much work. Hmm...
Also, to be totally honest, I have no intention of ever populating this circuit w/a 3207. Just a personal choice. Nothing against the 3207. I've explained my reasons for the 3007 choice elsewhere.
All the Best!
Dave

[moosapotamus]

This should read:
If using the MN3007 chip jump P to P1, R to R1, and T to T1.
If using a 3207 chip, jump P to P2, R to R2, and T to T2.
If anybody claims to have proofed this layout and didn't notice this: Shame on you!!! 

Doh! Just wanted to see who was paying attention. Um... you passed.

Also,(and I noticed this right before I clicked "post") the 1k R in the V divider is hardwired to ground as shown. For the 3207 setup this needs to be lifted from gnd so it can go to Vdd (pin 5, 3207). The 14k R to ground in the 3207 setup is correct. Connecting points R & R2 will result in both sides of the V divider being connected to ground. Might want to reroute some of them traces.

Ack! Right about that, too! I'll fix that tonight. Thanks!

I have to agree that the 78L09 reg should have some provision for filter caps. Better safe than aggravated later. Maybe moving C33 and that trace under it "south" just a bit would free up enough room for a 1u at the input and a .1u at the output of the reg? Might be kind of tight. But doable.

Well, there's already a 1u on the +15V rail (C32). So just a 0.1u on the +9V rail somewhere near C33 shoul do it, right?

I think it's a good idea to include the 3207 option as it would probably make alot of folks very happy. Plus, it's (almost) already done! On paper it should work. However my one concern; if it works; is how is it going to sound? ...

Absolutely. But in addition, I think Zben3129 raised a really good question, too... Does the 3207 need a positive clock frequency instead of the negative colck frequency used by all the others (MN3010, SAD1024, MN3007)? Kinda makes sense because logically it fits with the idea of having to switch the power and ground connections for the 3207. But how to do that? I don't really know... Didn't really want to get into having to change the clock circuit, too. Anyone else have an educated opinion on that?

Thanks
~ Charlie

[oldschoolanalog]

Doh! Just wanted to see who was paying attention. Um... you passed.

The only thing I passed is, uh, nevermind... 

Well, there's already a 1u on the +15V rail (C32). So just a 0.1u on the +9V rail somewhere near C33 should do it, right?

(Personal "doh!" moment) Yes, this is true. That should do it.

Absolutely. But in addition, I think Zben3129 raised a really good question, too... Does the 3207 need a positive clock frequency instead of the negative clock frequency used by all the others (MN3010, SAD1024, MN3007)? Kinda makes sense because logically it fits with the idea of having to switch the power and ground connections for the 3207. But how to do that? I don't really know... Didn't really want to get into having to change the clock circuit, too. Anyone else have an educated opinion on that?

Nothing needs to be done. The clock circuit is a positive clock and works just fine. The 1024 operates on a positive clock. So does the 3207. Use your meter and check the clock pins of the 1024. ~+7.7V IIRC. This should work properly for the 3207. Remember, I changed nothing about the clock circuitry and the 3007 board works. Its the 3007 and 3010 (which I based the 3007 work on) that need to be setup so they "see" (I dislike terms like that but I don't know the correct term)  the (-) supply & clock they require. Thats why the Vdd & Gnd are switched for each other w/the MN30XX BBD's. Not the 32XX BBD's. The 32XX are setup so Vdd is Vdd; and Gnd is Gnd. Just like the 1024. Someone like Mark Hammer or Stephen could explain the technical aspect of this better than me. I just know how to do it and that it works. 
More fun on the way!
Dave

[moosapotamus]

OK, I think that actually makes sense to me. So, once I get those other fixes done it should be good to go.

Thanks
~ Charlie

[Zben3129]

I see, I was assuming that sad1024 and mn3007 both used the same clock polarity and mn3207, but the mn3207 and sad1024 are the similar ones. And since this has been proven with a 3007 by simply reversing the clock output, then mn3007, mn3207, and sad1024 should all work.

What I was worried about was that if the mn3207 required the opposite polarity of the sad1024, that creating this would be impossible from the circuit. The obvious way would be to just reverse the outputs which woud theoretically work, but I wondered if doing this would create some kind of short or other internal problem in the clock IC causing failure. I assumed that since they make mn3101 and mn3102 that reversing outputs would be impossible, as why would they make two chips that are identical besides supply voltage and output polarity? Couldn't you just wire the mn3101 up in various ways to create all of the functions of the mn3102? The reason for the two chips must be just the difference in supply voltage and higher reliability of the mn3207. Or maybe reversing outputs on mn3101/2 is impossible, but since this circuit uses cd4047 (CMOS) as the driver it is possible.


Zach