TZF3007 - TZF (short delay) module, MN3007 based

[Pierre]

I'm building this TZF
http://thmq.mysteria.cz/TZF3007/v1_1/TZF3007_v1.1_Build_Instructions.pdf

...and this EM :
http://www.4shared.com/document/xNi5MG7d/Mistress_MN3007_SendRet.html

Btw. TDELAY range on all pictures here is wrong, correct range is 0.25ms ~ 4.5ms (TDELAY = 1024/(2*fCP))

...agh ! I don't undesrtand...

[Thomeeque]

Btw. TDELAY range on all pictures here is wrong, correct range is 0.25ms ~ 4.5ms (TDELAY = 1024/(2*fCP))

...agh ! I don't undesrtand...

Sorry, I have just findout that all pictures in this thread contain wrong information, so I have used reply to your questions as the "host" for this correction note (meant for everybody), it was probably not the best idea T.

[Puguglybonehead]

BTW; this circuit also can also be used as a nice & ez comb filter. Perhaps add a regen loop for that use. Put the delay pot on a pedal. Whatever...

Or build two, and have TZF controlled by a pedal....   

Hmmm.... Would that work? I'm tempted to try this sometime in the near future. Wondering which version of the board would be the better choice? I have both MN3007s and MN3207s hanging around here. Would I need to have a splitter of some sort in front of them and a mixer afterwards, or could the two boards simply be be wired directly to the input/output jacks?

[jdub]

Sorry for resurrecting this older thread, but I'm curious as to whether anyone ever tried implementing this board?  I'm in the process now of trying it with Tomas' EM3007 board (which btw sounds awesome)- I've got it wired in with a dpdt and there is a noticeable volume drop when the daughterboard is engaged, and also the effect is rather subtle.  I can hear the delay time changing when I turn the knob of the daughterboard, but the overall effect doesn't seem very present.  I remember Tomas mentioning that altering the value of the resistor in the return path of the dry signal (R25 in Markus' 9v EM schematic, R16 in Tomas' schematic and board) might help with "insertion loss".  In Markus' schem, R25 is 13k, while in Tomas' schem the value of this resistor is 10k.  In the opinion of the esteemed experts here, would lowering the value even more, to say 8k2 or even 7k5, help the volume drop in my case? 

Thanks in advance for any assistance!

[StephenGiles]

Sorry for resurrecting this older thread, but I'm curious as to whether anyone ever tried implementing this board?  I'm in the process now of trying it with Tomas' EM3007 board (which btw sounds awesome)- I've got it wired in with a dpdt and there is a noticeable volume drop when the daughterboard is engaged, and also the effect is rather subtle.  I can hear the delay time changing when I turn the knob of the daughterboard, but the overall effect doesn't seem very present.  I remember Tomas mentioning that altering the value of the resistor in the return path of the dry signal (R25 in Markus' 9v EM schematic, R16 in Tomas' schematic and board) might help with "insertion loss".  In Markus' schem, R25 is 13k, while in Tomas' schem the value of this resistor is 10k.  In the opinion of the esteemed experts here, would lowering the value even more, to say 8k2 or even 7k5, help the volume drop in my case? 

Thanks in advance for any assistance!

Why don't you connect up a 4k7 fixed resistor in series with a 4k7 linear pot and try?

[jdub]

I was planning on doing just that, just wanted to make sure I was headin' in the right direction before I start unsoldering stuff...

[jdub]

Well, I feel foolish  ...turns out that the problem was that the 2N3904 needs to be rotated 180 degrees; as currently shown in the module layout, Vcc is connected to the emitter and output is taken from the collector, and things should be vice versa.  This is clear from the schematic but I simply went by the layout and didn't check against the schem.  Volume is now good, although I need to tweak a bit to get the correct delay time from the module; effect is still not terribly strong.

So, anyone trying this, be aware that there is apparently an error on the layout.  Nonetheless, thanks to Tomas for his work in the first place!

[Mark Hammer]

I keep saying this.  You'd think that with only 3 pins, there would only be 9 different pinouts, but you'd be wrong.  There are at least 537 different TO-92 pinouts, some of them brought here by space aliens, and intended to stymie our technological progress so that they can continue to dominate us and use us for both food and crowd-scene extras in the humorous commercial they film here for consumption (along with man-meat) back home.

[StephenGiles]

I keep saying this.  You'd think that with only 3 pins, there would only be 9 different pinouts, but you'd be wrong.  There are at least 537 different TO-92 pinouts, some of them brought here by space aliens, and intended to stymie our technological progress so that they can continue to dominate us and use us for both food and crowd-scene extras in the humorous commercial they film here for consumption (along with man-meat) back home.

[jdub]

You seem to know an awful lot about these aliens, Mark...hmmm...

[cortezthekiller]

I had made a TZF flanger a while back by putting a small clone chorus in the send/return loop of the electric mistress EM3007 project circuit.
This allowed the through zero effect to be either a stationary point (with the chorus LFO off), or modulating along with the flanger LFO.

I have also finishing up a TZF with this project board along with the EM3007 project. There are two very important factors which need to be addresses if you want
to achieve the most dramatic through zero effect.

1. Both delayed signals must be out-of phase from each other. This project has an emitter-follower output which gives no phase inversion with respect to the flangers modulated delay line. In order to remedy this I had removed the components the signal is going through between the output of the MN3007 and the emitter of the output transistor. I had then reworked the component values and placement to change the output path to travel through a booster with the output on the collector side (I used an LPB1 circuit in this case). This allowed the signal to be out of phase from the other delay line and boosted enough so it could be balanced out to cancel the most signal (see point 2). (In this project there are 2 capacitors from the NPN collector to ground which must also be removed or else the signal will be shunted to ground).

2.Both signals should be the same amplitude (volume) to cancel out the most. Adding a 50k trimmer to the output to balance out the signal in filter matrix mode at the delay time that lowers the signal the most is important.

[jdub]

Wasn't aware of this...would you be willing to share your component values? Did you use the tzf board or an external board for the booster?  Thanks!

[cortezthekiller]

I used the TZF board from this thread. As I mentioned in my previous post I removed the output buffer/low pass filter circuitry and replaced it with an LPB1 booster (schematic can be easily found). This required some extra holes to be drilled to accommodate the biasing of the NPN transistor in that setup, though it was all able to be implemented fairly easily.

[jdub]

Cool, thanks...that's what it sounded like but wanted to be sure.  Sounds easy enough 

[jdub]

Hey Cortez- so I worked up the LPB-1 for the tzf path and I'm getting distortion on the output.  Did you change the gain factor of the LPB when you did yours?  Did you keep the volume pot/trim at the end of the LPB or just take full output?  Thx

[cortezthekiller]

I kept the volume pot after the LPB and had it as a trimpot. It wont need much volume to reach unity with the other delay line, which is what you want. So the trim acts as a mix for the two signals.
If you did not have the volume pot after the boost, the signal would be way too loud to allow the two to cancel each other out properly.

[jdub]

Righto- thanks.