EHX - Echoflanger  [schematic]

Discussion regarding early stompbox technology: 1960-1975 Please keep discussion focused and contribute what info you have...
User avatar
GodSaveMetal
Resistor Ronker
Information
Posts: 299
Joined: 19 Sep 2009, 16:11
Has thanked: 138 times
Been thanked: 55 times

Post by GodSaveMetal »

nelson wrote:Thanks Ron! Mind if I put that document on my site?

I've had it in mind to update this project for some time.

It was actually the pedal that gave me the impetus to pick up this hobby.

Since I'm returning to the hobby after a long hiatus.

It seems somehow poetic to make this one a priority to work on.

I no longer have the original working documents. So, it would be a bit of work.

I'm currently thinking about doing a project version with the MN3007. Since the SAD1024 is rarer than a virgin birth.

First thing that comes to mind in a conversion is the second SAD1024 is ran in parallel making it a 512 stage bbd. So, I'd need to double the clock speed then divide it to feed the other delay line in a MN3007 version. Since it is strictly a 1024 stage BBD. Even then, I wonder if the magic is somewhat due to the parallel 512 stage. So, it might be worth it to use a third MN3007 just to recreate it.

This thread and your build should be enough to make any errors/corrections clear to create a definitive (readable) Schematic. Which would be step one.

Any errors or things you pick up on. Please do post them here Govmnt_lacky!

I'll see if I can get Analogkid to chime in with what he remembers.

Thanks for taking this bad boy on!
Ohh goddness I´m waiting for that conversion for years!!! Mr. NELSON I be tuned on that! thnaks a lot man; in my country the MN3007 is 1 dollar cost!!! very cheap!! please post the conversion!!!!!!!!!!! :applause: :applause: :applause: :applause:

User avatar
Govmnt_Lacky
Information

Post by Govmnt_Lacky »

GSM,

I do believe that the dual SAD1024 chips is what kept this build in the basement cellar for so long. If an MN3007 conversion could be accomplished, then it would most likely be on everyone's to-do list!
The conversion would be daunting as, like stated above, one of the SAD chips is run in a parallel configuration however, they both share the same clock :slap:

So, unless you can split the clock signal to compensate for a dual MN3007 setup... I see no other way than using 3 of the MN3007 chips :shock:

I am far form an expert on these types of circuits (just enough to be very, very dangerous!) however, maybe someone will dive into this and attempt the conversion.

Cheers 8)

User avatar
Govmnt_Lacky
Information

Post by Govmnt_Lacky »

nelson wrote:Any errors or things you pick up on. Please do post them here Govmnt_lacky!
Other than what I posted above, there appears to be no other errors :D

Before I even printed a transfer, I compared the factory schematic, layout, and pics of the original unit. The only errors I found were posted here.

I will definitely post back with results and thanks to Ron, I should be able to get this beast tuned good.

User avatar
Dirk_Hendrik
Old Solderhand
Information
Posts: 4193
Joined: 03 Jul 2007, 08:44
Location: Old Amsterdam
Has thanked: 232 times
Been thanked: 887 times
Contact:

Post by Dirk_Hendrik »

OK,
For the "conversion to MN3007" wankers, again, it's a matter of doing it. No special magic involved. Stop begging for a schem and start studying dammit. A MN3007 can be used in place of a SAD1024. Setback is reversing the supply power from negative to positive ground (easy) and having a double delay time.

That means, as Nelson pointed out, doubling the clock rate.

In this case there's 3 SAD sections. 2 in series, one in parallel. That means that the 2 in series can be replaced with one MN3007 after which you're done.

But not for the parallel section. No need to parallel 2 MN3007's there. The only difference "could" be a somewhat different S/N ratio. For the better or the worse btw.
But that parallel section still needs to be half the delay time of the preceeding MN3007 (the one that replaced the 2 series connected SAD sections).

So the setup:
- Double the clock rate
- Divide the clock rate by 2 and use that clock for the first MN3007. (Hint: A 4013 dual flipflop will do that for you cheap and effective)
- Use the double clock for the other (2nd) MN3007.

done.

The above too hard? Start studying.

@ Nelson,
Wether those files you posted years back are verified and correct or not, They sure did help a lot me 2 years ago troubleshooting my echoflanger. Saved me frome endless flipping over the board. Thanks for that!~!~
Sorry. Plain out of planes.

http://www.dirk-hendrik.com

User avatar
Fender3D
Cap Cooler
Information
Posts: 573
Joined: 08 Aug 2009, 02:40
Location: Helltaly
Has thanked: 121 times
Been thanked: 153 times

Post by Fender3D »

Govmnt_Lacky wrote:...
So, unless you can split the clock signal to compensate for a dual MN3007 setup... I see no other way than using 3 of the MN3007 chips :shock:
Halving the clock won't be difficult since you have still half 4013 free...

User avatar
Govmnt_Lacky
Information

Post by Govmnt_Lacky »

Dirk_Hendrik wrote:OK,
For the "conversion to MN3007" wankers, again, it's a matter of doing it. No special magic involved. Stop begging for a schem and start studying dammit. A MN3007 can be used in place of a SAD1024. Setback is reversing the supply power from negative to positive ground (easy) and having a double delay time.

That means, as Nelson pointed out, doubling the clock rate.

In this case there's 3 SAD sections. 2 in series, one in parallel. That means that the 2 in series can be replaced with one MN3007 after which you're done.

But not for the parallel section. No need to parallel 2 MN3007's there. The only difference "could" be a somewhat different S/N ratio. For the better or the worse btw.
But that parallel section still needs to be half the delay time of the preceeding MN3007 (the one that replaced the 2 series connected SAD sections).

So the setup:
- Double the clock rate
- Divide the clock rate by 2 and use that clock for the first MN3007. (Hint: A 4013 dual flipflop will do that for you cheap and effective)
- Use the double clock for the other (2nd) MN3007.

done.

The above too hard? Start studying.

@ Nelson,
Wether those files you posted years back are verified and correct or not, They sure did help a lot me 2 years ago troubleshooting my echoflanger. Saved me frome endless flipping over the board. Thanks for that!~!~
Dirk,

First and foremost... Thank you for this contribution.

Second... Some of us (me included when it comes to altering circuits) have to start somewhere. I realize that there were a few requests earlier in the thread for someone to "Come up with a solution" however, realizing my short comings in this area, I requested assistance as I am sure MANY, MANY others have done in the past. I am not looking for someone to do ALL of the work for me (at least not without some form of compensation) but it is always nice to get a helping hand from those who are more knowledgable. At least until others (myself included) have studied and caught up.

Now... I cannot speak for others but, I am extremely appreciative for the help I have received and I attempt to pass that along to others with problems that I am familiar and knowledgable on. This just happens to be an area that I am not "caught up" in my studies on.

I just want to thank Dirk, Ron, Nelson, Federico, and everyone else that has helped :thumbsup

Nothing but love for all of you :hug:

User avatar
nelson
Breadboard Brother
Information
Posts: 53
Joined: 28 Aug 2007, 20:45
Has thanked: 2 times
Been thanked: 10 times

Post by nelson »

Dirk_Hendrik wrote:OK,
For the "conversion to MN3007" wankers, again, it's a matter of doing it. No special magic involved. Stop begging for a schem and start studying dammit. A MN3007 can be used in place of a SAD1024. Setback is reversing the supply power from negative to positive ground (easy) and having a double delay time.

That means, as Nelson pointed out, doubling the clock rate.

In this case there's 3 SAD sections. 2 in series, one in parallel. That means that the 2 in series can be replaced with one MN3007 after which you're done.

But not for the parallel section. No need to parallel 2 MN3007's there. The only difference "could" be a somewhat different S/N ratio. For the better or the worse btw.
But that parallel section still needs to be half the delay time of the preceeding MN3007 (the one that replaced the 2 series connected SAD sections).

So the setup:
- Double the clock rate
- Divide the clock rate by 2 and use that clock for the first MN3007. (Hint: A 4013 dual flipflop will do that for you cheap and effective)
- Use the double clock for the other (2nd) MN3007.

done.

The above too hard? Start studying.

@ Nelson,
Wether those files you posted years back are verified and correct or not, They sure did help a lot me 2 years ago troubleshooting my echoflanger. Saved me frome endless flipping over the board. Thanks for that!~!~

Glad my work helped you out!

Yeah, the S/N ratio is the only consideration. I was kinda floating the idea based on slight variations between the two internal BBD's creating slightly diff sounds. Which is a silly idea. Given that the differences would be negligible because they're both in the same package. So, 1 MN3007 at 2x the clock rate would be fine. Other half of the 4013 to do the frequency division.

Doubling the clock rate should be simple enough. They use a capacitor switch to change from 'chorus' to flange clock ranges. I never really noticed that they also switch out the second output filter when upping the clock speed to flange ranges.

The Filter Matrix and Slapback modes just switch out the LFO for steady comparator voltage and change the frequency range of the clock with the 150pf + 47pf Cap to ground off the LM311. A pot here would let you futz with the delay time....

I wish I could devise a way to get rid of the CD4053. However, even removing the superfluous Blend(A hardware SPST to switch an electronic SPST) and the bypass electronic switching. I'd still need more throws than a standard lorlin rotary switch could provide. The best I can think to do would be to simplify it to one CD4066. Which has the four switches needed. This would remove one IC from the board.

The mod creep will irritate you Dirk but I don't care. The two BBD's would make adding a TZF mod *really* tempting. O, and a psuedo stereo mod while I'm at it.

User avatar
nelson
Breadboard Brother
Information
Posts: 53
Joined: 28 Aug 2007, 20:45
Has thanked: 2 times
Been thanked: 10 times

Post by nelson »

nelson wrote:
Dirk_Hendrik wrote:OK,
For the "conversion to MN3007" wankers, again, it's a matter of doing it. No special magic involved. Stop begging for a schem and start studying dammit. A MN3007 can be used in place of a SAD1024. Setback is reversing the supply power from negative to positive ground (easy) and having a double delay time.

That means, as Nelson pointed out, doubling the clock rate.

In this case there's 3 SAD sections. 2 in series, one in parallel. That means that the 2 in series can be replaced with one MN3007 after which you're done.

But not for the parallel section. No need to parallel 2 MN3007's there. The only difference "could" be a somewhat different S/N ratio. For the better or the worse btw.
But that parallel section still needs to be half the delay time of the preceeding MN3007 (the one that replaced the 2 series connected SAD sections).

So the setup:
- Double the clock rate
- Divide the clock rate by 2 and use that clock for the first MN3007. (Hint: A 4013 dual flipflop will do that for you cheap and effective)
- Use the double clock for the other (2nd) MN3007.

done.

The above too hard? Start studying.

@ Nelson,
Wether those files you posted years back are verified and correct or not, They sure did help a lot me 2 years ago troubleshooting my echoflanger. Saved me frome endless flipping over the board. Thanks for that!~!~

Glad my work helped you out!

Yeah, the S/N ratio is the only consideration. I was kinda floating the idea based on slight variations between the two internal BBD's creating slightly diff sounds. Which is a silly idea. Given that the differences would be negligible because they're both in the same package. So, 1 MN3007 at 2x the clock rate would be fine. Other half of the 4013 to do the frequency division.

Doubling the clock rate should be simple enough. They use a capacitor switch to change from 'chorus' to flange clock ranges. I never really noticed that they also switch out the second output filter when upping the clock speed to flange ranges.

The Filter Matrix and Slapback modes just switch out the LFO for steady comparator voltage and change the frequency range of the clock with the 150pf + 47pf Cap to ground off the LM311. A pot here would let you futz with the delay time....

I wish I could devise a way to get rid of the CD4053. However, even removing the superfluous Blend(A hardware SPST to switch an electronic SPST) and the bypass electronic switching. I'd still need more throws than a standard lorlin rotary switch could provide. The best I can think to do would be to simplify it to one CD4066. Which has the four switches needed. This would remove one IC from the board.

The mod creep will irritate you Dirk but I don't care. The two BBD's would make adding a TZF mod *really* tempting. O, and a psuedo stereo mod while I'm at it.
Ok, the CD4066 would work if it weren't for the fact they're SPST switches. :oops:

User avatar
nelson
Breadboard Brother
Information
Posts: 53
Joined: 28 Aug 2007, 20:45
Has thanked: 2 times
Been thanked: 10 times

Post by nelson »

Here's a redraw of the schematic.

Couple of values I am unsure on. They have question marks next to them. I can't remember how rigorous I was in verifying every component all those years ago.

I'd appreciate if people cast their eyes upon it and saw if they could spot any glaring errors that my eyes missed.

I'm totally not making a new PCB for the SAD1024 version. I'm just using the schematic as a basis for modifications.

I thought I'd post it here as a more readable version for folks future repairs/reference.
Attachments
EFRD.pdf
Echoflanger Schematic Redraw
(67.37 KiB) Downloaded 405 times

User avatar
Govmnt_Lacky
Information

Post by Govmnt_Lacky »

Nelson,

I went over your original files and listed the discrepancies in my previous post. I matched ALL of the component values with the EH schematic and the only problems were listed.

After comparing your schematic with the original EH schematic, I see a few missing and incorrect values. When I get more time, I will try to list them but here are a few:

1) BBD null needs to be a 1K trimmer
2) Missing a 2.4K resistor in parallel with C30 on your schematic (R15 on the EH schematic)
3) C35 on your schematic should be .047uF instead of 0.18uF (this is C18 on EH schematic)

Hope these help!

User avatar
Govmnt_Lacky
Information

Post by Govmnt_Lacky »

OK,

After comparing schematics, this is what I have come up with in addition to the changes above:

1) D2 and D5 are 1N4001 diodes.
2) D3 needs to be a 5mm RED DIFFUSED LED. I believe this is crucial for providing the correct Vbb to the SAD chips.
3) C18 is a 1n cap.
4) C24 is a 2.7n cap

These are only the unlabeled values and the values that have questions marks.

Cheers :D

User avatar
nelson
Breadboard Brother
Information
Posts: 53
Joined: 28 Aug 2007, 20:45
Has thanked: 2 times
Been thanked: 10 times

Post by nelson »

Govmnt_Lacky wrote:OK,

After comparing schematics, this is what I have come up with in addition to the changes above:

1) D2 and D5 are 1N4001 diodes.
2) D3 needs to be a 5mm RED DIFFUSED LED. I believe this is crucial for providing the correct Vbb to the SAD chips.
3) C18 is a 1n cap.
4) C24 is a 2.7n cap

These are only the unlabeled values and the values that have questions marks.

Cheers :D
Yeah, I left the diode type unlabelled as they're clearly 1N400x types.

Thanks, I have probably missed more. I'm not sure whether R25 off pin 5 of the NE570 is 47K or 9.1K. 9.1K would fit better with other designs utilising the chip. However, it really does look like 47K.

User avatar
Govmnt_Lacky
Information

Post by Govmnt_Lacky »

nelson wrote:
Govmnt_Lacky wrote:OK,
Thanks, I have probably missed more. I'm not sure whether R25 off pin 5 of the NE570 is 47K or 9.1K. 9.1K would fit better with other designs utilising the chip. However, it really does look like 47K.
It is 47K (R2 on EH schematic)

User avatar
nelson
Breadboard Brother
Information
Posts: 53
Joined: 28 Aug 2007, 20:45
Has thanked: 2 times
Been thanked: 10 times

Post by nelson »

Here's the MN3007 version schematic.

Bypass + Blend switching still as per original.
Attachments
Echoflanger MN3007 V1-1.pdf
(65.67 KiB) Downloaded 260 times

User avatar
nelson
Breadboard Brother
Information
Posts: 53
Joined: 28 Aug 2007, 20:45
Has thanked: 2 times
Been thanked: 10 times

Post by nelson »

Not allowed to edit posts?

Anyway, ignore previous schematic. Biasing for MN3007 wrong.

Here.
Attachments
Echoflanger MN3007 V1-1.pdf
(66.17 KiB) Downloaded 352 times

User avatar
GodSaveMetal
Resistor Ronker
Information
Posts: 299
Joined: 19 Sep 2009, 16:11
Has thanked: 138 times
Been thanked: 55 times

Post by GodSaveMetal »

nelson wrote:Not allowed to edit posts?

Anyway, ignore previous schematic. Biasing for MN3007 wrong.

Here.
GREAT Mr. Nelson !! IS THE IC10 AND IC11 MN3007??? thank a lot!!!!!

User avatar
nelson
Breadboard Brother
Information
Posts: 53
Joined: 28 Aug 2007, 20:45
Has thanked: 2 times
Been thanked: 10 times

Post by nelson »

GodSaveMetal wrote:
nelson wrote:Not allowed to edit posts?

Anyway, ignore previous schematic. Biasing for MN3007 wrong.

Here.
GREAT Mr. Nelson !! IS THE IC10 AND IC11 MN3007??? thank a lot!!!!!

Nah, they're a Reticon R5101.

User avatar
Fender3D
Cap Cooler
Information
Posts: 573
Joined: 08 Aug 2009, 02:40
Location: Helltaly
Has thanked: 121 times
Been thanked: 153 times

Post by Fender3D »

Strange..
They look like TDA2107..






:mrgreen:

User avatar
phibes
Transistor Tuner
Information
Posts: 1107
Joined: 10 Dec 2008, 01:17
my favorite amplifier: HIWATT motherfucker!
Location: United States
Has thanked: 75 times
Been thanked: 189 times
Contact:

Post by phibes »

Image
GuitarlCarl - "TGP = The Gear Polishers"

Image Ken

User avatar
nelson
Breadboard Brother
Information
Posts: 53
Joined: 28 Aug 2007, 20:45
Has thanked: 2 times
Been thanked: 10 times

Post by nelson »

This thing is a bitch to route.

Post Reply