• 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 freeLog in
    Want zero display ads or expanded classifieds tools? Compare plans.

SWR Grand Prix - frequency analysis of preamp and EQ

alexclaber

Commercial User
Jun 19, 2001
5,264
3,024
Brighton, UK
www.barefacedbass.com
Disclosures
Director - Barefaced Ltd
Right, I've finally done it! I thought it would be difficult but it was a piece of cake. Downloaded some cunning software that a the very clever fdeck had written, installed it, plugged output of soundcard into input of preamp and output of preamp into input of soundcard. I highly recommend doing this, it's very very enlightening. Let me know if you want to know links to software and more in depth info on how to do it.

The software lets you paste the data into excel so you can then create comparison graphs. It would be great to build up a database of these so you can understand what the 'GK' or 'SWR' sound is and how to EQ another amp for it.

Bear in mind that the scale varies from graph to graph, as does the relative gain.

C%3A%5CDocuments%20and%20Settings%5Caclabe%5CMy%20Documents%5CMy%20Pictures%5CSWR%20GP1.jpg


So here we can see that the default SWR Grand Prix tone (with this particularly NOS 5751 valve) has a broad mid scoop centred at 250Hz and about 8dB deep compared to the bass and treble peaks. Also that boost in the lowest lows explains why SWR amps are incorrectly referred to as being underpowered - they simply put out too much in the region where the cabinet cannot reproduce it. The broad treble boost explains the inherent bright clear tone. I was surprised that it was this far from flat as when I've AB'd between preamp and through FX loop (which I scoped and found was near perfectly flat) with the bass it's sounded very very similar. However when I've played program music through the amp by plugging in an iPOD or CD player it's always sounded somewhat shy in the low mids. This can be pretty well corrected using the semi-parametric EQ to add about 8dB of boost at about 250Hz.

C%3A%5CDocuments%20and%20Settings%5Caclabe%5CMy%20Documents%5CMy%20Pictures%5CSWR%20GP2.jpg


Interesting point here is that the bass and treble EQ seem to be peaking not shelving EQ. I'll try and get round to doing full cut plots.

C%3A%5CDocuments%20and%20Settings%5Caclabe%5CMy%20Documents%5CMy%20Pictures%5CSWR%20GP3.jpg


Here's the big myth destroyer! What you can see on this plot is that the aural enhancer barely does anything to the EQ below 2 o'clock and then by the time it gets to max it is a deep high Q cut at about 250Hz. What you can't see is that as you turn up the aural enhancer it adds gain until about 2 o'clock, in a not insignificant amount. I understand the second half of the dual triode tube in the preamp is used for the aural enhancer so I presume what we're hearing is another load of preamp tube gain, which presumably adds a dash of tube character - it certainly doesn't change the EQ as SWR claim in the lower regions and in the higher regions it's just a simple mid cut (plus a gain boost to compensate).

Alex
 
Peaking rather than shelving is more logical, there's no point in boosting EQ either below or above where the speaker can actually do anything with it. If anything it appears that this rig is voiced a good half octave lower than it should be for best result, but appearances can decieve, since the response of the amp must be added to the response of the speakers to get the end result. If the speakers have a response bump at 100 Hz, which is likely, that would explain why the amp isn't voiced to do the same.
 
This is outstanding!!! Can you post a link to the software and detailed setup instructions? I've got 5 preamps I can test.

Its funny, when I use the grand prix I run it pretty much flat other than a big boost around 250hz! I should just use a strait gain box :bassist:
 
Here's Francis J. Deck's homepage:

Invalid Link Removed

This is what you want:

"My free real-time analyzer software: Useful for measuring preamp voicing and speaker response on a Windows PC"

It's really awesome! Download this file: Invalid Link Removed

Unzip it and open the folder to run the install. The pdf help file is pretty good but in summary you simply connect your preamp to your soundcard click on the noise generator and analyze tick boxes, and adjust the input gain on preamp and output volume on computer for best match, and then output volume on preamp and input gain on soundcard so the level bar is green. Voila, your first plot. Click on the paste to clipboard button to get the numbers out and then paste them into excel. Repeat until you've got all your measurements and then make your graphs.

I really want to see how other preamps shape up!

Alex
 
Trace Elliot SMX dual compressor ... engaged

Now when someone asks what punch looks like we can just use this graph. :bassist: Sweet. How much compression were you using?

I'd beg everyone who posts on this thread to be as thorough as Alex was. Record all of your gain stage choices and settings when posting a curve. That will make the most of your time and the thread.
 
I really want to see how other preamps shape up!

I'll do a few I get a chance. I use Rightmark Audio Analyzer though: http://www.rightmark.org/, which is freeware and also does distortion analysis, signal/noise measurements, and some other cool stuff. I like fdeck's program, but I don't have Excel on my computer so it doesn't make sense for me. But the ability to watch things in real time is a real bonus with fdeck's app.
 
I'll do a few I get a chance. I use Rightmark Audio Analyzer though: http://www.rightmark.org/, which is freeware and also does distortion analysis, signal/noise measurements, and some other cool stuff. I like fdeck's program, but I don't have Excel on my computer so it doesn't make sense for me.

Well, if you run the program and can output a text file I can turn it into a graph for you (that goes for anyone who wants to do this but doesn't have Excel). PM me if interested.
 
unfortunately my excel skills are zero, so i can't efficiently use fedck's software.

the results from the rightmark software are easier to read, i hope.

what is apparent: the markbass features a boost in the very low end when all knobs are centered.

and there is no evidence of the perceived 250 hz bump that somem have heard.

and: from what i have seen in fdeck's realtime software, the controls have caracteristics that differ quite a bit from the graphs in the manual.