Guyatone BB-1 Flip Valve 'Bass Driver'
Guyatone BB-1

Currently residing on my workbench is an old Guyatone BB-1 Flip Valve “Bass Driver” pedal, on loan from OzBass forum user Crayzeebass. It’s fairly heavily influenced by the BK Butler / Chandler Tube Driver, (made famous by a Mr David Gilmour of a little band called Pink Floyd you may have heard of) and runs on 9 volts. He’d purchased it from another forumite, but commented to me that it lost a bit of his tone. I’d been curious to try it out (it has a neat semi-parametric mid control), and he’s local, so he dropped it off to me for to have a look-see.

First up, let me say this: it’s an awesome little tube distortion pedal for guitar – if you see one cheap, GET IT. Upon plugging one of my Precision basses into it and running it through my Fender Pro rig (TB600 head and 215 Pro cab), it does seem to lose some bass frequencies. I’d always wanted to do a proper frequency analysis of some of my preamps & pedals, so this seemed the perfect opportunity.

First of all, you’ll need the following:

  • PC with USB recording interface (I use a Line 6 Toneport)
  • Headphones (plug into the monitor output of recording interface)
  • Multi-track DAW software (I use Audacity, free download)
  • Spreadsheet software (Excel, Open Office Calc)
  • Pink noise soundfile (just Google for one)
  • 1/8″ stereo to 1/4″ mono adaptor (from PC line out to guitar cable)
  • 3′ guitar lead (from PC to unit being tested)
  • 6″ patch lead (from test unit to USB interface)
  • Something to test – FX pedal or preamp, mic’d guitar cabinet

A couple of quick notes: it’s possible to test passive circuits (and pickups!), but not with this method exactly. I’ll do a followup post later, as I’ve been meaning to do some further investigations regarding passive tone controls, cable length and suchlike. Also, through repeated tests I discovered that my resolution is limited to +/- 0.2 dB – your hardware may be better or worse, the only way to find out is to perform the following test on the same audio several times and compare your results. I won’t do anything like this again until I’ve updated some of my hardware!

Safety note: Pink noise is usually at or close to the 0 dB mark – ie, LOUD. It’s not essential to listen to the samples being played, so I keep the headphones off (or around my neck) so I can hear the audio is coming through okay.

Audacity pink noise
Pink noise

First of all, open Audacity & import your pink noise file. At this point I save the file with an obvious name, like “BB-1 pedal test”. You then need to run the Line Out from your PC straight into your USB interface. The purpose of this is to re-record the pink noise and establish a base level frequency profile for your particular combination of soundcard, cables and USB interface. Adjust your playback and recording levels until your recording is not clipping, but is as close to the original pink noise file as possible. Once you’re satisfied, make a recording of this new (pink) noise, and label the track ‘Soundcard’ or something similiar.

Step two is to hook up the pedal or preamp in question, then re-record your original pink noise file with the pedal bypassed.  Another obvious label will come in handy; if you’re testing one pedal, label it ‘Bypass’. You already have enough data to make a meaningful statement about the effect the pedal is having on your signal chain. It was at this point I decided I was going to test the bypass mechanisms of a bunch of different pedals, so I quickly hooked them up in various combinations and blasted some pink noise through them.

Spectrum analysis
Spectrum analysis

In Audacity, select the ‘Soundcard’ clip, then open Analyze > Plot Spectrum. This will generate a funky purple graph – leave the Algorithm as ‘Spectrum’, and Function as ‘Bartlett Window’. Adjust the Size to 2048 samples or higher, as this gives us enough resolution to cover the low end, and make sure the Axis is ‘Log frequency’. You can then Export this data, which will give you a text file of Frequency (Hz) vs Level (dB) measurements. Open this in your spreadsheet as a tab or space delineated file, and you should have two seperate columns.


Having fun yet? Then let’s continue! If you haven’t already guessed, you then repeat this Export process for your ‘Bypass’ recording (and any others), and bring the new data into your spreadsheet. Remove subsequent Frequency (Hz) columns, as you only need it once for your  X axis. Column A contains the Frequency, Column B contains the Level data for your Soundcard measurements, and Column C onwards contain the Level data for your bypassed pedal(s). Most of your data will be in the negative scale, so you’ll need to use a simple equation to generate some nice lines:

Relative level = (-1 * B) – (-1 * C)

Relative levels

Relative levels

So, for the first row of data at A3, Level = (-1 * B3) – (-1 * C3). Simply enter the equation into a free cell at say, G3, then copy and paste that equation down the column. It is then a simple matter* of graphing Column A (Frequency, Hz) as the X axis (logarithmic) versus Column G (Relative level, dB), as the Y axis (linear). If your graph looks a little squashed, you can adjust the X axis to only show data from 20 Hz to 20kHz, and the Y axis probably only needs to go from -9 dB up to 9 dB unless you have a serious boost pedal to test! Consider also focusing on a smaller picture – if you’re worried about lost low end, then restrict your data to 400 Hz or lower, for example.

* Unless you’re having the same issue with labelling the X axis in Open Office Calc that I am! In that case I highly recommend downloading Datascape from Select the columns you want to graph, hit Copy, then Paste into a blank txt file in Notepad. Save as, then open the file with Datascape.

I ended up testing the following:

  • Boss buffered bypassPedal bypass methods
  • SansAmp buffered bypass
  • True bypass
  • Three linked true bypass pedals
  • Three linked Boss pedals

The graph shows the results, which I find quite interesting. That’s a Log scale on the X axis from 10 Hz to 20 kHz, apologies for the bad labelling – I had issues with Open Office Calc, but Datascape is not as good at showing this much information. First of all, please note that I appear to have hit that snag with resolution I mentioned earlier, so any differences less than 0.2 dB should be taken with a grain of salt. You’ll notice straight up that the single True Bypass signal is actually very slightly above unity gain – this is obviously a spurious result (probably caused by the cheap gear I’m using!), but the other characteristics of the line are still useful – note that it’s pretty flat in the lower end.

There’s a big difference between the SansAmp buffer, the Boss buffer and the mechanical but non-true bypass BB-1 switching. The SansAmp buffer has none of the crazy harmonics of the Boss and hardly any signal loss – it basically looks like a true bypassed signal. The Boss buffer is not at unity gain, and stacking three of them accentuates both the low- and high-end rolloff. The BB-1’s vintage bypass clearly has to go – it’s nothing but insertion loss & tone suckage! Next step is to get my hands on a decent Tube Screamer, (it’s supposed to have a good buffer as well), and a selection of other common pedals to test.

Once you’ve got the process down pat, you can repeat it with the subject pedal(s) on different settings. This way you can compare what’s going on as you change the amount of distortion, or adjust the EQ. This is particularly handy for pedals or pre-amps with poorly marked controls. For example, if you know a parametric Eq has a mid range from 100 Hz to 2.5 kHz, you can find out what frequency is boosted at the halfway point and determine if the control is linear or logarithmic.

For those with preamps that use the passive Fender tone-stack (also Alembic, Marshall, Vox etc), you can get a better handle on what the controls are actually doing to your sound – you might be surprised. In this case, you can also measure amplifiers with a pre-amp out or FX-send section – just make sure you’re not running a tube amp with no load, okay?!

Oh, and for those of you wondering about the Guyatone BB-1, it’s a counter-intuitive pedal: the clean signal that can be blended back in has LESS bass than the distorted signal with the bass boost engaged. Go figure 🙂 I’ve been commissioned by CrayzeeBass to mod the pedal, so it will be the subject of a follow-up post. This may seem like a lot of effort to go to, but not all of us have the ears of a recording engineer. Knowing what’s happening to your audio at various points along your signal path will help you get the best tone possible, and that’s something we should all strive for.