I am designing a tube preamp for bass guitar that performs as a balanced DI for sending the signal to the board, or as an output for a power amp. I am starting from the output and working backwards so that I get my gain stages right. My design calls for an output stage to be a 12au7 triode as a cathode follower into a 4:1 output transformer (9600 to 600 ohms). The output audio would be 5 volts stepped down to 1.25 volts (+4dBu). This balanced signal would be used either as a DI or as the signal to a power amp for personal monitoring. Is this a good approach for the design goals? Are there any compromises that I'm introducing? Do you have any advice for achieving such a goal?
I prefer to leave substantial headroom over the nominal power amp sensitivity spec, but I tend not to want the sound of tubes driven hard. You might take a look at the Jensen website and peruse their line driver circuit for the 10K61 transformer. If you can swing another tube the Aikido circuit sounds great for bass applications IME. I use 6SN7s in mine, but 12AU7 should be fine too.
You'll need a huge output transformer to handle the low end without core saturation at 0dB plus headroom (don't assume that the lowest fundamental being about 40Hz means there won't be information lower than this), but that shouldn't be a problem (apart from weight)
Don't forget to put a ground lift switch if you're using it as a DI, and, if you're going to be doing festivals and the like, add a switchable 20dB attenuator after the output transformer; not all PA desk balanced mic inputs are happy about true line level. Actually, a separate transformer winding at lower voltage/impedance is better, because then you can split off passively to PA FOH mixer, monitor console, recording truck, TV truck… without hitting any loading problems, and keeping your personal stage system, but unless the transformer is specially wound it's difficult to manage this.
But the principle's good, and should sound good, too.
Don't forget to put a ground lift switch if you're using it as a DI, and, if you're going to be doing festivals and the like, add a switchable 20dB attenuator after the output transformer; not all PA desk balanced mic inputs are happy about true line level. Actually, a separate transformer winding at lower voltage/impedance is better, because then you can split off passively to PA FOH mixer, monitor console, recording truck, TV truck… without hitting any loading problems, and keeping your personal stage system, but unless the transformer is specially wound it's difficult to manage this.
But the principle's good, and should sound good, too.
Thanks for the posts...the output transformer is a dilemma for me. I'd like to fit the pre into a 1U rack, so size is a consideration. The Jensen DI does fit, but would need to be mounted horizontally, perhaps against the back panel. The other transformer I am looking at is the Cinemag CM-2810. The cinemag comes in a pcb format and the datasheet looks good, but I wondered about the size of it. If its an air core, it can produce the low frequencies easier without saturation, right?
According to the datasheet, it has an iron/nickel core and would stand approximately 1,25" from the circuit board.
For an audio transformer to have an air core, it would likely need to be huge.
For an audio transformer to have an air core, it would likely need to be huge.
Last edited:
The amount of wire needed to get an air core down to 20Hz doesn't bear thinking about. Air cores are for radio frequencies and crossover inductors.
No, a good traditional iron cored is the answer, and quality wise, nobody can complain about the Jensen, but it will be big. and heavy. No way round it with the low end and the level (well, an audio toroid might cut down on the weight a bit – if you could find one the right ratio).
No, a good traditional iron cored is the answer, and quality wise, nobody can complain about the Jensen, but it will be big. and heavy. No way round it with the low end and the level (well, an audio toroid might cut down on the weight a bit – if you could find one the right ratio).
It may or may not fit depending on which particular 1RU case you use, measure carefully. It does sound quite nice in my experience. I just happen to be finishing up a similar build today in 2RU format, using 2-6SN7 and one 6SL7, and the Jensen JT-10K61-1M:
Here's an older thread on another forum on a 3RU build I did using one of John Broskie's Aikido boards.
Last edited:
Do you use your Jensen loaded pre as a DI to a mixer, or as a signal to a power amp?
Another question for all: Does the current drive capabilities of the output triode come into consideration when running a balanced signal for either a DI or a signal to power amp?
Both ways, sometimes into an EFX in on other bass amps, and occasionally also straight to soundcard for recording. My primary use is driving an IcePower 1000ASP equipped plate amp for live gigging on fretless bass guitar though.
Excellent. That's the kind of application I'd like to get out of this design as well.
Hello fellow bassplayer 🙂.
Is there a specific reason you want to use a transformer for the ballanced output? Why not use a 12AU7 in a cathodyne phase splitter configuration?
Is there a specific reason you want to use a transformer for the ballanced output? Why not use a 12AU7 in a cathodyne phase splitter configuration?
Good to see some bass players on this forum 🙂
The way I understand it, I would want a high impedance load on the output of the preamp. If the output impedance of a cathode follower 12AU7 is ~450 ohms, the input impedance of the mixer/amplifier would need to be stepped up as they are not inherently large. Would this not aid in driving long balanced cables, such as in a 500 ft snake?
Remember, while a cathode follower can pull up with all the current it can pull through the tube it can only pull down with the cathode resistor, so the impedance might be low, but the actual drive current available is asymmetrical. Into a higher impedance, or with lower excursion, this is irrelevant, but on an active balanced output, I suspect the two sides of the balanced not being total mirror images.
Besides, a nice lump of iron in a DI gives me confidence in its separation, and your preamp's going to have lots more volts kicking around in it than most line inputs would like to meet, if the thing is ground lifted (I know, I know. All stages have perfect grounding, and I've never had to check round with a neon screwdriver to see which mics were 'hot' relative to which amps).
Besides, if you are forced to a lesser transformer the distortion will probably sound quite pleasant; transformer distortion, in small quantities generally does.
Besides, a nice lump of iron in a DI gives me confidence in its separation, and your preamp's going to have lots more volts kicking around in it than most line inputs would like to meet, if the thing is ground lifted (I know, I know. All stages have perfect grounding, and I've never had to check round with a neon screwdriver to see which mics were 'hot' relative to which amps).
Besides, if you are forced to a lesser transformer the distortion will probably sound quite pleasant; transformer distortion, in small quantities generally does.
Is this the characteristic of the ever elusive "tubey" compression sound that guitarists/bassist look for when chasing different tube amps? I'm not looking for hifi, but I'm also not looking to botch my signal either. I admit I haven't implemented this yet, but I definitely want a coloured tone, like that of the vintage variety (Fender, Sunn, Ampeg to name a few).
While I agree with the need for a transformer in DI application to provide the isolation, safety, etc. I don't think the above statement is correct, since the current flowing through the plate and cathode of a triode is exactly the same, and when the plate and cathode have the same resistance on them as in most cathodyne designs, their outputs should be exactly the same (into identical loads of course).
In the case of a guitar/bass amp, when the cathodyne is over-driven itself, then the balance will be thrown off as the grid current begins to flow on the output tubes. But it is hard to see that happening in the DI + soundboard application, since the soundboard typically has fully balanced transformer and/or opamp inputs. Of course, with long cables, grounding, EMI, etc. there may still be some load im-balances, but that would be the case whether a cathodyne is used or not.
Jaz
AFAIK the 'old tube compression' is the result of carbon resistors. They tend to run hot when current passes through them. Hot carbons have higher resistance. So resistance constantly varies depending on the current being drawn. Voila, compression.
Last edited:
I think you are talking about the extreme cases only, the compression from tube amps in general DO NOT come from resistors, carbon comp or otherwise. 😛
Correct, and in small amounts off course, not a substitute for tube compression caused by e.g. non linearities or DC coupled stages, but guitarist seem to prefer these carbon filled amps.
FWIW In my tube preamps I originally planned on using "iron" balanced transformers, but as an interim solution made a little PCB with a TL072 + a switchable attenuator (+4 and -20 dBV).
Works so good and reliably that I left it as-is forever and kept using it in following ones.
This one has 4 +4dB unbalanced to drive 2 stereo rack PA amps (in this case a QSC900 and an AB1500 into 4 Ampeg 8x10") + the XLR balanced one to send signal to the Stadium mixer 60 meters away (because of zig-zagging actual wire path was closer to 100 meters).
Sorry for the messy dirty shop, maids aren't what they used to 😉
To have an idea of the Stadium size across which those balanced signals were sent without a hitch:
By day , slowly filling up:
(say Hi to my Daughter Victoria)
by night, roughly in the middle of the show:
I doubt that a simple 12AT7 cathode follower would have driven that long balanced cable so easily as the humble TL072, even with a transformer.
I think that to beat it with a pure tube solution, it would be needed at least a 12AT7 used as a small power amp (similar to Fender Reverb drivers) or a small power tube, such as a 6AQ5 or similar.
Check 50's and 60's classic Radio and Recording line drivers.
Last edited:
Hello Victoria!! 🙂
That's the exact same solution I used for my DI, I just didn't dare to say I used sand in my pre 😛. Works perfect. I made the attenuator adjustable by pot, but either way is fine off course. Frequency respons is way better compared to iron for the same size and cost. And less coloration for that matter. I've never had to run those lengths, but good to hear it's capable of doing so.
I still used a cathodyne for the power amp output though.
That's the exact same solution I used for my DI, I just didn't dare to say I used sand in my pre 😛. Works perfect. I made the attenuator adjustable by pot, but either way is fine off course. Frequency respons is way better compared to iron for the same size and cost. And less coloration for that matter. I've never had to run those lengths, but good to hear it's capable of doing so.
I still used a cathodyne for the power amp output though.
Last edited:
- Status
- Not open for further replies.