• TalkBass has been independent since 1998. Add your voice.
    Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
    Join free Log in
    Want zero display ads or expanded classifieds tools? Compare plans.

TB11 Preamp Project

#93
A couple of constructive questions:

1. The B-B OPA1641 appears to be optimized for this application (excellent performance over audio frequencies, low quiescent current, great unity gain performance), any reason why you didn't use it? UPDATE - found this in the older posts, just based on current, the OPA2277 might be a better choice if your bass is still powered by batteries. I power mine through a TRS cable.

2. With each pup buffered by an opamp, you don't need an active blend stage to preserve tone. Could you eliminate the middle buffer? With the tone controls in the feedback loop of the last opamp, the passive blend shouldn't interfere with the tone controls. With a quad chip though you might as well use the extra opamp if its there.

3. Has anyone SPICE modeled the tone control circuit? If the frequency bands overlap, the controls will interfere with one another. The Baxandall circuit works, in part, because there's no overlap - there's no mid and the hi/lo frequency bands are place far enough apart that there's negligible interaction. If this were considered a problem, you may need a separate mid control (a better use for that last opamp). UPDATE - I found the posts where you talk about the simulation results but the graphs and figures don't load - still wondering about interaction.

4. Someone else mentioned the lack of DC blocking caps.

5. Another TB'er (rumblinbass - an electronics design engineer) likes to put small caps in various and not-obvious places around opamps to improve stability. I can probably dig up his suggestions from the other thread if you're interested.
 
#94
Rob22315 said:
A couple of constructive questions:

1. The B-B OPA1641 appears to be optimized for this application (excellent performance over audio frequencies, low quiescent current, great unity gain performance), any reason why you didn't use it? UPDATE - found this in the older posts, just based on current, the OPA2277 might be a better choice if your bass is still powered by batteries. I power mine through a TRS cable.
Click to expand...

That looks like a good sub if you can live with the higher current (the slew rate increase might be worth it).

2. With each pup buffered by an opamp, you don't need an active blend stage to preserve tone. Could you eliminate the middle buffer? With the tone controls in the feedback loop of the last opamp, the passive blend shouldn't interfere with the tone controls. With a quad chip though you might as well use the extra opamp if its there.
Click to expand...

Probably.

3. Has anyone SPICE modeled the tone control circuit? If the frequency bands overlap, the controls will interfere with one another. The Baxandall circuit works, in part, because there's no overlap - there's no mid and the hi/lo frequency bands are place far enough apart that there's negligible interaction. If this were considered a problem, you may need a separate mid control (a better use for that last opamp). UPDATE - I found the posts where you talk about the simulation results but the graphs and figures don't load - still wondering about interaction.
Click to expand...

This is a Baxandall topology, however bass only spans perhaps 1KHz of fundamental frequency. It's exceedingly difficult to put in a midrange that doesn't interact somewhat with the other bands. Not sure this is a problem however (read post #33).

4. Someone else mentioned the lack of DC blocking caps.

5. Another TB'er (rumblinbass - an electronics design engineer) likes to put small caps in various and not-obvious places around opamps to improve stability. I can probably dig up his suggestions from the other thread if you're interested.
Click to expand...

From post #61:

One thing that was omitted (on purpose) was pole compensation and DC blocking stages. I've found that keeping the phase shift to a minimum tends to 'sound better', and makes applying global negative feedback (not used in this design) easier to implement. I'm guessing that with low GBW op-amps that are low passed to around 40kHz that oscillations aren't going to be a problem, but we shall see in the prototypes.
 
#95
MglMatador said:
This is a Baxandall topology, however bass only spans perhaps 1KHz of fundamental frequency. It's exceedingly difficult to put in a midrange that doesn't interact somewhat with the other bands. Not sure this is a problem however (read post #33).
Click to expand...

I'll read through #33 carefully later today but it would be easy enough to move the mids to a separate opamp if there's an extra one on the quad chip.

I'll dig up rumblinbass' other posts and repost them here then you can make your own judgement.
 
#98
I actually haven't forgotten about this, but alas getting slammed at work for four solid months leaves little time for fun projects. ;)

I made some preliminary layouts for the initial incarnation of the TB11 preamp. I split the design into two parts: the first part contains the power supply rail splitter, the capacitive loading switch (one for each pickup), the low-Z switch (one for each pickup), and the input impedance resistors (what would be the volume pots on a passive bass). It contains two unity gain buffers and pretty much nothing else: it's probably good enough as a complete project all by itself for someone who doesn't want controllable frequency bands and just wants clean impedance conversion for their passive bass:

tb11_frontend.jpg


Despite making routing a real PITA, I left the DIP8 footprint in so that opamp substitutions would be possible with a DIP8 socket. I provided enough clearance to mount one. I used radial caps for the PS, but 1206 surface mount resistors to save space. I just couldn't shoehorn in even 2mm body vertical resistors without running out of clearance.

Board size is 1.5" x 0.8", however I used this size only because I have a few small gaps on a multi-project PCB I am putting together, so thats the size I used. I could probably get this on a 1" square board provided I remove some of the mounting holes and switch to an SO-8 package for the opamp (which saves a LOT of space, but makes assembly harder for those who aren't used to soldering SMD components).

Here is the second board, which contains the filter (it's the same size, just the scale printout was slightly different):

tb11_filter.jpg


I wanted to use polypropylene film caps so I paid a hefty layout penalty. Resistors are 1206 as well, however I had to put a handful of them on the bottom of the board. My next multi-project panel should have bigger spaces so I can do a more proper layout.

I went ahead and removed the active mixer section, and moved the mid control to the other opamp section. Thus the first opamp buffers both pickups, which are passively summed into the second stage, which contains the bass and treble pots. The output of this stage then feeds the midrange control. I added layout provisions for caps to be placed into the feedback loop in case there are high frequency oscillations. As soon as I get the schematics cleaned up I'll post both of them.

I measured the output offset on my prototypes, and since gain is low (maximum voltage gain through the entire chain is 5 at maximum boost), output offset voltage does not seem to be a problem (and these BB OPA2277's seem to perform very well in this regard). Thus there are no interstage AC coupling caps in this design. It helps that everything is using bipolar supplies so that the non-inverting inputs can be ground referenced.

I'm doing an alternative layout which is 1" x 3" to hopefully be able to use only PTH components. I think 1" x 3" is small enough to fit in just about any bass cavity (even the narrow jazz routing), but if I can manage to make it smaller I will.
 
#99
And just as I was re-reading it. Good to hear you're back, sir. :D Being an EE and circuit design dullard, I have to ask something.

If I got this right, the boards you made are not exactly to the schematic reposted here.

  • V+/V- splitting, cap-and-Z switching and individual buffer used to be on the first two schematic segments in the PDF, they are now crammed together on the first board.
  • The active summing used half the op-amp, the tone control the other half on the third schematic in the PDF. Now you have passive mixing instead of the active summing, the bass and treble use half the op-amp and the midrange control has its own half.

Pardon my ignorance - but what is the benefit of the midrange having its own half of preamp? Would it, in theory, enable you to make a semi/fully parametric (freq or Q) midrange control?
 
#100
Stealth said:
And just as I was re-reading it. Good to hear you're back, sir. :D Being an EE and circuit design dullard, I have to ask something.

If I got this right, the boards you made are not exactly to the schematic reposted here.

  • V+/V- splitting, cap-and-Z switching and individual buffer used to be on the first two schematic segments in the PDF, they are now crammed together on the first board.
  • The active summing used half the op-amp, the tone control the other half on the third schematic in the PDF. Now you have passive mixing instead of the active summing, the bass and treble use half the op-amp and the midrange control has its own half.

Pardon my ignorance - but what is the benefit of the midrange having its own half of preamp? Would it, in theory, enable you to make a semi/fully parametric (freq or Q) midrange control?
Click to expand...

You got the bullet points exactly right.

The benefits of having the mid control separate are several:

1) Makes design modular. If all you want are bass and treble controls, you don't need to stuff components in the last op-amp section. If all you want is a mid, then you can do the same with the bass and treble sections. The mid can also be made parametric much easier if it's separate.

2) Lessens interactions between the bands. As was said above, a useful mid control might interfere with the other controls (depending on where they are placed in the frequency spectrum). Interacting controls can cause strange phase problems and odd hills/valleys in the frequency response.

3) Makes analysis much more tractable. The three band shelving EQ section ends up having half a dozen poles/zeroes in the solution making analysis of the controls difficult.
 

Latest posts

Top Bottom
Back