• 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.

Build your own tube-preamp

I wonder if there is any interest in homebrewing a bass tube preamp. I´m currently designing-building a 12ax7 tube preamp.
Schematic and simulated frequency response in the attachement.

I'm currently building the preamp. Any comments/advice on cross over frequencies etc. is welcome

BRs

Willem
The Netherlands
 

Attachments

A lot of us here love a good DIY thread; Passinwind and myself already have a few preamps under our belts. Always cool to see more.
 
Cool! Looking forward to seeing your construction details, including what opamp you're using. I generally prefer mids centered considerably lower than 1K, but that's easy enough to change to taste. Maybe consider buffering after the second tube stage as well once you are using opamps?

Here's the thread
on my last DIY build. A new, more compact version will be done in a couple of weeks and I'll post construction pictures as I go.
 
Thanks for your comments guys!

Reading through your preamp projects made me decide to make some major changes to the tone controls. Now I have:
- parametric low section (30-120Hz, Q=0.5...2 )
- parametric mid section (150-600Hz, Q=0.5..2 )
- passive high (1.5kHz)
- low notch/low off switch.
This makes a nice versatile tone control.

Opamps: not decided yet. Might use LM4250.
I have also added an opamp as a final stage, to obtain
the desired low-Z output. (Good hint).

I finished the preamp power supply today. It is made of two cheap 2*6.3V/18W PCB transformers in cascade,
which gives me 300Vdc, 12Vac and +/- 8Vdc dual supply in one go. PCB is 4" x 8" (homebrew).

I also bought a nice aluminum case on ebay today, with a nice 0,2" thick aluminum front panel.
The front panel is just big enough for all the controls/switches.
Guess this is the most expensive part of the preamp.

I will report progress in this thread.

BRs

Willem
Netherlands
 
In my totally biased opinion, you aren't going to get the full monty with a solidstate buffered output. Transformers are the way to go, but those are more difficult and expensive. (but way better than autobots!)
Why use op amps at all? It is supposed to be a "tube" pre?

Why build a tube preamp and run it into a solid state amp with an opamp in the first stage? ;) That said, I'm all for using output transformers when budget and space constraints allow, even with solid state preamps.

There are other possibilities too, including using a high voltage MOSFET like this guy is. He'll be posting a thread on TB as his build progresses.
 
After some (re)search for a full tube design I found a useable design which had a 3-fold tone control in feedback loop.

Did some mods, so that it would match with my requirements.
Unfortunately not yet with parametric mid/low section, but
I finally want to start building, since my new preamp case was delivered last Saturday.

Schematic and frequency response posted for comments . .

BRs

Willem
 

Attachments

Hey Willem,

Cool design. Whats the value of R32 and R36? Seems like 1e8 could mean 100M (which isn't practical) or 1.8ohm (which also isn't practical). Did you mean 1.8K?

Also, I'd try to put the volume control before the output buffer. I think an AC coupled cathode follower instead of the DC coupled version you have there would allow you to put the volume control right after the second gainstage.
 
@Stromrider
The 1E8 (100M) resistors have no other purpose than a DC path for the simulation software.
I agree it's confusing and I should have deleted them from the schematic.

As for the gain control, I was thinking to replace R41 and change it for a 20K log pot.
That would give a control range -4dB to 12.5dB (overall gain). X17 is to be replaced by a 270k resistor.
The benefit is that I can maintain the DC coupling of the input stage, which reduces the number of capacitors in the signal path.

Any thoughts on this?

Willem
Netherlands
 
I think X17 (the gain control on the latest schematic) is fine as is. That should give you plenty of control over the drive to the second gainstage (X5).

My comment was directed at the volume control X10. I would make X11 an AC coupled cathode follower, and put the X10 volume control after the X5 gainstage. It would require adding two more caps, and a couple resistors for biasing X11, but I think it would be worth it. It would be especially valuable if you ever drive a modern power amp with an input impedance down around 10Kohms.
 
@Stromrider
The 1E8 (100M) resistors have no other purpose than a DC path for the simulation software.
I agree it's confusing and I should have deleted them from the schematic.

As for the gain control, I was thinking to replace R41 and change it for a 20K log pot.
That would give a control range -4dB to 12.5dB (overall gain). X17 is to be replaced by a 270k resistor.
The benefit is that I can maintain the DC coupling of the input stage, which reduces the number of capacitors in the signal path.

Any thoughts on this?

Willem
Netherlands

They don't affect the input and output impedance of the tone stack then?

My first worries looking at that is in practice your bass and treble knobs are to high and wont really get the effect you're looking for.

Also designs like this don't normally lend themselves to clipping very well if you were to be going down that route aswell
 
I will have to do some further research as how I can achieve sufficient low output impedance, so that this will not influence the tone control.
I was thinking it is on the safe side (<50k), because the cathode followers will give relatively low output impedance.
At least the controls of this tone stack have less mutual influence than the standard Fender tone stack.

With respect to the center frequencies, these can be modified quite easily.
I have chosen following 3dB points: 150Hz for low, 350Hz for mid and 1.0kHz for high.
With the given component values this should give a modest control range (plus/minus) 12dB at 31Hz, 7dB at 350Hz and 9dB at 10kHz.

For the homebrewers I have include the pcb design for the preamp.
(Fat resistors = 1W / 0.5" pitch). I will post the tone control pcb with the pots later

BRs

Willem
Netherlands
 

Attachments

Looking good :) I see you're doing a proper job of it, PCB's and all.

I've been working on my own tube preamp using 6SL7 tubes. First prototype is almost done, second is in progress. I am waiting for a push/pull pot to arrive so I can get rid of the switch hanging out the bottom. The circuit is based on an early Ampeg B15N with a lot of tweaks. It's all point-to-point handwired, easy to tweak and change parts, which I have done a LOT as I have been "voicing" it.

I tried making a plexiglass case for the second, but it was a PITA to work with, so I am building it into a nice 1U rackmount case.

I like your design, using DC coupled cathode followers to reduce the number of caps and so you don't need an output transformer.

From my experience, most modern power amps can handle a wide range of impedence. I had a homemade output transformer in my prototype, but it didn't have enough bandwidth, rolled off both highs and lows, so I took it out. I'm using the 2nd tube as a phase inverter and running it through a dual 1M attenuator to a balanced XLR out and works pretty well with my Crown power amp.
 

Attachments

  • 6SL7_pre_wood_1.jpg
    6SL7_pre_wood_1.jpg
    37.7 KB · Views: 824
  • 6SL7_pre_wood_2.jpg
    6SL7_pre_wood_2.jpg
    75.6 KB · Views: 568
Looking good :) I see you're doing a proper job of it, PCB's and all.

I've been working on my own tube preamp using 6SL7 tubes. First prototype is almost done, second is in progress. I am waiting for a push/pull pot to arrive so I can get rid of the switch hanging out the bottom. The circuit is based on an early Ampeg B15N with a lot of tweaks. It's all point-to-point handwired, easy to tweak and change parts, which I have done a LOT as I have been "voicing" it.

Way cool! Looking forward to hearing some clips of it. Once you have everything else close enough that you're tweaking the voicing the real fun is at hand.

I was planning on doing a thread on printed circuit board design and fab after I finish up my current DIY builds, but I am far from an expert in this. Willem, could you talk a little about that?
 
I will have to do some further research as how I can achieve sufficient low output impedance, so that this will not influence the tone control.
I was thinking it is on the safe side (<50k), because the cathode followers will give relatively low output impedance.
At least the controls of this tone stack have less mutual influence than the standard Fender tone stack.

With respect to the center frequencies, these can be modified quite easily.
I have chosen following 3dB points: 150Hz for low, 350Hz for mid and 1.0kHz for high.
With the given component values this should give a modest control range (plus/minus) 12dB at 31Hz, 7dB at 350Hz and 9dB at 10kHz.

For the homebrewers I have include the pcb design for the preamp.
(Fat resistors = 1W / 0.5" pitch). I will post the tone control pcb with the pots later

BRs

Willem
Netherlands

I'd recommend doing a PTP version before you launch into making pcbs! Computer simulations are great but they often don't translate into real world situations super well.

Looks good and should be a good project though
 
thank you all for your responses so far. Very helpfull.

@passinwind and all other homebrewers, see below a description my well proven protoype PCB development method:

I use MicroCap software for schematic design, simulation and analysis. Once the schematic design is ready, I design my prototype pcb.

I use Sprint Layout 5.0 pcb design software for simple pcb's. This software is very intuitive, however it does not provide schematic support, rubber banding etc. It has a nice components library and editor (so that you can draw your own components and footprints).

For prototyping I use the 'toner method' to transfer my layouts to the pcb. (This method does not works without a decent laser printer) Once the prototype design in Sprint Layout is ready I print the mirrored design on standard glossy foto paper.

Then comes the tricky part: I have to wait until my wife is out of the house. At an unguarded moment I get hold her electric iron and I iron the layout on a clean and lightly sanded piece of circuit board. Put the printed side of your layout on the copper surface, heat up the iron at max. heat and push firmly, with the iron at max temperature (no steam). After a minute or two, the toner melts and the paper sticks to the pcb. The photo paper may discolor lightly.

Let the pcb cool down a while and put it in a small tank/bowl with water. After cooling down, put the iron back where it originally came from, without letting your wife know you used it for other purposes :)

After 30 mins or so in the water, the paper is saturated and it loosens from the pcb. Peel it off carefully. Use an old toothbrus to brush all remaining paper off. The toner remains on the pcb (when you have done things right). If you spot any damaged traces, you can try and repair with waterproof ink (in EU: use an edding 8400 pen). If traces have damaged too much, the toner transfer was done at wrong temperature. You just clean your pcb and start with a new transfer. Not too many costs, except lost time. . .

Send your wife away (again... she better not knows... :)

Put on old clothing, rubber handgloves and protective glasses.
Check pcb for a 2nd/final time. Heat up a water pan in your kitchen and make sure your kitchen is (well) ventilated.
Put a jar with etchant in the water pan and heat it up until the water just boils. Then turn of the heat. Drill a small hole in your pcb, put a piece of isolated wire through and put your pcb in the jar. Let the etchant dissolve all uncovered copper. Normally this is ready in 10-30mins, so check the etch process with regular intervals (every 2 mins) by taking the pcb out of the etchant and expose it to the air. Do not inhalate over the etchant (too much :)

After etch process has finished, clean with plenty of water.
Use a 'metal sponge' to rub away the toner or use some fine sandpaper. Clean again. Drill pcb. Clean again. Prototype PCB ready.

= = =

If you're not in to all this chemist work, you may as well order your pcb at a professional pcb manufacturer. Export the gerber files from your sprint layout design and send them to the manufacturer.
(This will probably take some more time and costs to realise a prototype pcb. . .)

Hope this helps.

= = =

Below I have a picture of my preamp enclosure. (bought on Ebay). I will post pictures of the built process within next weeks.

BRs

Willem
Netherlands
 

Attachments

  • preamp case.JPG
    preamp case.JPG
    15.8 KB · Views: 407
@passinwind and all other homebrewers, see below a description my well proven protoype PCB development method:

<snip>

If you're not in to all this chemist work, you may as well order your pcb at a professional pcb manufacturer. Export the gerber files from your sprint layout design and send them to the manufacturer.
(This will probably take some more time and costs to realise a prototype pcb. . .)

Hope this helps.

Thanks very much for the detailed answer. To get my feet wet I used KiCad for board layout (open source freeware, expect the Y in DIY to assert itself at times) and Oshpark.com for board fabrication. Oshpark sells only in multiples of three @ $1.66 per square inch per board, but they are fairly fast if you live in the U.S. and the quality is quite good. I have done a lot of rework on my new solid state preamp board this week with no issues at all. Here's what that board looked like after initial component stuffing:

PW7B_stuffed.jpg


My wife is totally hip to my tricks, I'll save subterfuge for other things....;)