TECH21 Sansamp BassDriver DI (or something like that) *SCHEM*

[George Giblet]

The original circuit uses some JFET to switch between clean and effect/ampsim.    If you look at the PCB pic posted in this thread you can see a HEF 40xxx (40106?) chip in the top right and to left of the black stuff you can see two JFETS.  That part of the circuit is similar to most commercial effects, and many schems on the net leave that part out. You can replace it with a mechanical switch.  You don't really want true-bypass on this one because it disables the balanced output - see notes below.

I haven't gone over the layout with a fine tooth comb but it definitely looks like it is based on the schematic.

There' a few things to note:
- The XLR circuitry has been removed, everything from R30 and beyond.
- It appears the layout is for a "lite" effects pedal version where the circuit is wired using a bypassing mechanical switch (which bypass the buffers)
- C17 is removed.  Which is part of the bypass/switch stuff.  You can see C20 is hard wired to C21 on the layout.
- The output resistor R28, at the output,  is removed.  It should be put back in.
- There is grayed-out part of the layout - not sure what that is about.  The electro cap appears to be C33.  The other stuff errr..

Regarding the schematic:  The JFET switching has been pulled out a little hastily.    Each of the caps C17, C20 and C21 should have 1MEG resistors wired back to Vb (I traced this from the pcb photo).  The resistors prevent popping when the bypass footswitch is pressed.  (When replacing the JFETs with a mechanical switch, arguably these could also be wired back to 0V.)  This has no bearing on the lite version.

I've perhaps elaborated too much on this so it sounds complicated the above issues are all trivial and easily resolved.

[George Giblet]

OK I've put together a simplified lite version which performs *identically* to the original but removes the XLR stuff.  Quite a few opamps can be removed.  The lite version is shown with two quad opamps but you could use 4xduals or 2xdual+quad..

http://www.geocities.com/george_giblet/effects/bd21_lite_v10.png

Note: The previous layout given is NOT for this lite version of the circuit.

[Kornell]

Thanks George!!

[Shakal]

Nice work man! Thanks a lot.

I'll take a better look later.

The original Bass Driver DI has 2 or 3 little black switches, but I can't read what are they. Do you know what are they for?

[George Giblet]

From what I can see:
- 6.5mm line level or instrument level
- XLR line level or instrument level
- XLR phantom/ground connect; a common thing for XLR.

The first two would probably switch the gain of the appropriate stages.

[Shakal]

So... no big deal in not having them?

I'm sorry, but I didn't get the "buffer bypass" thing. Would it be that ON/ON switch? If it is, there is a few parts of the circuit before the bypass, won't it interfer? It can be made with just a SPDT stompswitch ?

Sorry about the obvious questions... I'm quite newbie, and your knowledge is an opportunity to learn... I'm just taking it, if you don't mind, of course. 

[George Giblet]

The line/level switches are nice to have when you need them but you can survive without them. They add to the cost and construction (and are one less thing to go wrong).  No real use for the lite version.

The phantom power/ground lift is an XLR thing.  Not sure how much you know, or want to know, about XLR stuff.   Phantom power is where the thing you connect to supplies power via the audio cable, so you don't need batteries or power supplies at the remote box.  Ground lift is used to help some types of  ground loop problems, which produce humm.   Phantom power and ground lift are completely different things, however, if you lift the ground the phantom power stops working on a lot of equipment.    For an effects pedal both aspects aren't of much use, for the XLR stuff they are part of that world.  You can read up on it:

http://en.wikipedia.org/wiki/XLR_connector
http://en.wikipedia.org/wiki/Balanced_audio

Buffered bypass is where the signal runs through a buffer stage when the effect is bypassed/off (that would be U1:A in my circuit). The advantage of a buffer is the it provides a consistent load impedance to the thing connected to the input, this prevent loss of highs due to long cable runs or when connecting to low impedance inputs (like a sound card or mixer input).   It generally helps keep noise problems down as well.  The other alternative is true-bypass which completely disconnect the effect, so the box is literally wired through.  Neither is better as such but they do have differnt behaviours in different conditions, particularly when multiple effects, long cables, and low impedance loads are concerned.   For bass I think buffered bypass has the edge, for guitar perhaps true-bypass (mainly because some effects need to load the pickup).

The other type of bypass is "output bypass" or partial bypass.  In the this case when the effect is bypassed the effect input circuit is left connected to the input signal and output socket is also wired to the input signal.  With this method the effect adds to the load on the input signal.  If you have a string of effects like this the load can get quite low and affects the tone.  It's not a good method to use but was popular in older effects.  This problem is probably why there was a big movement towards true-bypass mods!

[Bernardduur]

Awesome!

Why are R10 and C10 altered??

[MartyMart]

Great work George, much appreciated :-)
I own a "real" one so wont be making it but the info is useful, sound guy takes my DI out for the
out front sound and I send the link to my bass rig, sounded great out front and no need for cab mic !

[George Giblet]

Thanks guys.

> Why are R10 and C10 altered??

To saved an opamp I moved the filter circuit C15, C16, R18, R21  from the output of the filter involving R10/C10 to the input side.  When this is done the two filters interract and the response is changed, *even though no part values are changed*.  To make the filter behave  like the original the R10 and C10 values need to be changed.  The result is a simpler filter which *behaves the same* as the original.  The match between old and new is extraordinarily good, within 0.1dB across most of the band with one point at 0.2dB deviation.  With capacitor tolerance this is closer than the unit to unit variation of originals in production.

Here's a comparison:
http://www.geocities.com/george_giblet/effects/bd21_filter_match.png

I should tell you getting the filter to match was a lot of work (and requires a fair bit of skill).  The fact only two part values changed R10 and C10 makes the task look a lot easier than it was - I was actually surprised how simple it turned out.  Trying to economize filters like this usually requires a lot of part changes to the extent that the filters no longer look remotely similar, even though they behave the same!

You can sleep easy that the unit will behave like the original.

[moosapotamus]

That's excellent work!

Thanks!
~ Charlie

[vanessa]

I have a question,

On the original it has a "parallel output". This comes in really handy when playing live (feed it to your amp and run the Bass DI to the house) or even in the studio to have one signal split to the board or to a mic'd amp.
I'm not seeing where they "tapped" it off in the schematic. Is there a buffer stage for it or is it just tapped off the input?

[George Giblet]

> "parallel output"

By tracing the PCB pics I'm quite certain the parallel out is simply wired to the input terminal ie. no buffering on the parallel out.

[vanessa]

Thank you! One more for the road,

If I were to cut out the cab simulator to use with say the ROG "Flipster" as the front end. Where is that located on the schematic? Is the sim in the upper right hand corner (Rx or Cx? to Rx or Cx?)?

I think that the Flipster might be a better front end if you're looking for that "Ampeg" tone. I've always used the Bass DI in the "Ampeg" mode and wished I could separate the cab sim from the pedal to put the Flipster in front.

[Shakal]

Perfect! Thank you so much George Giblet.

Now, working on the Layout. Does anyone knows a easy way to do it? I'll try with Multisim's Ultiboard, with no guaranteed success.

I'm just curious...  how is it going to sound with guitar?

[12afael]

a lot of time that I was not exited with a new circuit, I will begin tomorrow!
Thanx George

Rafael

[George Giblet]

> If I were to cut out the cab simulator to use with say the ROG "Flipster" as the front end.

It's basically the circuit from and including R16 through to and including C14,  you then need to add about 50k from the  C14 to ground.

You probably should remove simple speaker sim (the two RC networks) from the output of the flipset.

I suspect the BD21 speaker sim will load down the Flipster, which will also modify the BD21's cab filter.  Easiest to add a buffer.

I don't really think the speaker sim is has been tuned to match an ampeg, since the basic circuit is the same as other tech 21 products and they clamed to match quite different speakers!

The network R3/C5 through to R5 heavly modifies the response.  On the Tech 21 classic this claimed to produce a Bassman (?) emulation.

>Now, working on the Layout. Does anyone knows a easy way to do it? I'll try with Multisim's Ultiboard, with no guaranteed success.

There's a few free one around PCB express is another.  Most are about the same, it's just a matter of perseverance and familiarity.

[Shakal]

Let's work on it. 

[Bernardduur]

How could I add a mid knob to this??? I usually don't like the cut in mids on my bass pedals

[Shakal]

The non-existance of a mid controller doesn't mean that the mids are cutted.

I think it can't be so hard, anyway it's a good idea!

George, can you help us with it?

Ps: There is a pedal with mid controller, its the Sansamp Para Driver DI (very similar to Bass Driver)