Bass Amp Project

[Scott Anderson]

Here is my first tube amp project. It is based on the Dynaco MKIII and Sunn 200s. Bass Amp Project

[mesherm]

Nice job!
Just a suggestion to try.
If all your using the amp for is a bass guitar, try dropping your bias current to 100 ma or so (50ma a tube) and see if you notice any difference. I am running 6550s in triode mode in a home built stereo amp and I have them set at 40ma each and I don't notice any ill effects. Running your 6550s a little cooler might give you a bit more headroom. The worst that could happen is you would be running closer to class B.

[Shannon Parks]

Here is my first tube amp project. It is based on the Dynaco MKIII and Sunn 200s. Bass Amp Project

Hi Scott,

Great project! Thanks for the fine documentation and notes. I think that switching to Ned's driver board will make a difference for you, as you'll have a dedicated driver whipping those KT88's into shape. I would definitely do that before trying different tubes.

In the meantime, I wonder if you would be able to do some open loop gain and closed loop gain tests. Since this is a bass amp, maybe another 3dB negative feedback could be a plus - increasing the dampening and 'tightness' of the sound. You could try values between 500 and 1K for the feedback resistor and increase you feedback cap, too, so you don't get unstable. Without measurements, I guess would probably make it 2000pF. Basically I would size this as large as possible by ear.

Shannon

[Scott Anderson]

Thanks for your comments guys. I have a 2 year degree in electronics that I received in 1974. But I haven't really used it since the mid 80s. I'm now have a 4 year CS degree and work in the CS field. I use to do a lot of hobby stuff until then and have gotten rid of most of my stuff. The only equipment I have are a couple of soldering irons, an old DMM, a suspect pocket analog meter, an old hobby built 5 volt and 12 volt dc power supplies, an old hobby built frequency counter and various jumpers, parts and test leads. It's going to take a little time to get new test equipment and restock the parts bins.

Shannon, I did try Ned's board with a 12AU7 and 12BH7 combination but had mixed results. More noise than with the 6AN8 board. But have to admit that's it's hard to experiment with it installed in the chassis. I'm going to buy another set of transformers, some KT90s and do some breadboard testing before I do the final assembly. I'll retry Ned's board again using 2 - 12AU7s. When I start breadboard testing then I’ll try the changes to the feedback components as you suggested.

My main concern now is the 12AX7 preamp circuit. It has a lot of noise and when I adjust the bass and treble at certain points, the amp will squeal sending the dog running and barking. It will be easier to breadboard the preamp circuit. I can already bypass the preamp now and go straight to the 6AN8 board.

The amp sounds great now with the exception of the preamp. I have a Sansamp Bass DI that I can plug in to the input of the 6AN8 board. The KT88s are biased at 140 ma and my best guess is the amp is about 70 watts. To be honest, sounds louder than my 300 watt Mesa hybrid.

[dhuebert]

Firstly I like to say, it's what YOU like which is most important, disregard any of the following if it's not right for you...

With bass amps, damping factor is important. Something I did to lower output impedance by half was to double up on the output tubes. You are running ultra-linear (UL) (Local Negative Feedback LNF) which is good but I would say ditch Global Negative Feedback (GNF). From what I can see you have no bypass capacitors on the cathodes of the preamp stages, this is LNF too, which, again, is good. The use of LNF in all the stages reduces or eliminates the need for GNF. Something I will try in my next project, which is a 6V6 push pull amp is 0.005 uF caps between the cathode and anode of the output tubes to limit high frequency response (LNF). I built a few amps and played with GNF, sweeping the signal generator to 100KHz and beyond and found no instabilities, just reduced power with GNF.

Why in heaven's name would you use tube rectification? That is strictly for guitar amps that want power supply sag for increased sustain. For a bass amp you want a tight, stiff (read low inpedance) power supply to deliver high power attack and low distortion sustain. LF4007, fast recovery sand for modern power supplies.

Take out the 10 ohm resistor in the cathodes of the KT88s and replace it with a 1 ohm for each cathode to ground. You are dissipating power in that 10 ohm and it's not really needed, at all. Talk to a guru for recommended bias voltages and take the resistor out alltogether.

Usually, in a guitar amp the tone shaping occurs in the tone controls and the interstage coupling caps. Do you need tone controls? Get rid of them if you don't. They introduce phase distortion which is good for high distortion guitar amps but bad for bass amps. Try increasing coupling caps for more bass.

If you haven't already, read:

http://www.tone-lizard.com/
http://www.normankoren.com/Audio/index.html
http://www.webace.com.au/~electron/tubes/

All great sfuff.

Don

PS:
http://www.diytube.com/forumpix/power_amp.jpg

http://www.diytube.com/forumpix/power_supply.jpg

http://www.diytube.com/forumpix/preamp2.jpg

[Scott Anderson]

dhuebert, thanks for your comments. Like I said this is my first tube amp. Heck my first tube anything. The dust is very thick on my diploma and as I blow that dust off I’m discovering lots of rust. It’s been 20+ years since I’ve messed with electronic stuff and 30 years since I messed with tubes. In fact I only messed with tube circuits when in school or replacing them in my amps.

I’m taking a guess here, but when you talk about LNF and GNF are you talking about the resistor, cap coming from the output transformer and KT88 grid. If so, which is which? You can see the rust can’t you?

Why a tube rectifier? Well it was on the schematic. The Sunn 200s was a bass amp. Many of the old bass amps used tube rectifiers. Yes it may be a little dirty but being an old school bass player dirt is sometimes nice. This amp one was for learning or let’s say, relearning. I pretty much followed the 2 schematics to the letter. My next amp project may be different.

I’m going to try removing the 10 ohm cathode resistor soon. I’m not sure how to measure tube bias without it.

Do you need tone controls? Yes, tone is what a musician wants and strives for. I can’t imagine an amp without it. It would make a player one dimensional. The tone knob on the guitar is not enough in my opinion. I’ve seen some players remove the tone control from the instrument to use only the tone shaping from the amp. In fact some players remove all the controls on their instrument and let the amp control their sound.

Thanks for the links. I’ll read and learn from them as I can. I’ve learned a bunch of stuff here at diytube

This stuff is fun to do. But man is it expensive and I have a lot of stuff to buy.

[Shannon Parks]

My main concern now is the 12AX7 preamp circuit. It has a lot of noise and when I adjust the bass and treble at certain points, the amp will squeal sending the dog running and barking. It will be easier to breadboard the preamp circuit. I can already bypass the preamp now and go straight to the 6AN8 board.

The amp sounds great now with the exception of the preamp.

Hi Scott,

Forget all that tweaking stuff Don & I were spouting. Your amp has a little gremlin and we can get it purring before you hot rod it. The squealing part is a major red flag there is something wrong, and we can fix it.

1) Apart from the A, B & C voltages, did you measure the cathode and grid voltages? If you have'em, post them here for us.
2) If you can also email me some super close up shots (in macro mode) of the interior, that would be great, too.
3) Could you list any deviations from the original schematic?
4) Lastly, does it start squealing when treble is maxed then you slowly max the bass, or vice-a-versa, or either way?

Shannon

[dhuebert]

It has a lot of noise

What kind of noise? Hissing (thermal noise) Humming? (power supply problems) Buzzing? (grounding problems) Squealing? (feedback phase problems).

Do you need tone controls? Yes

I'm new to guitar amps too, but the guitarists I've spoken to have said no to tone controls. I guess this is where I have alot more learning to do.

Why a tube rectifier? Well it was on the schematic

With UF4007s and a switchable sag resistor you can control from clean to dirty with a flip of a switch.

I’m not sure how to measure tube bias without it.

No, you're right. In my schematics you see a 1 ohm resistor to measure current. Old school builders didn't bother with this instead using experience to set bias. Something we're short on!

More later.

Later: GNF is anything that spans multiple stages, whereas LNF is confined to one stage. So the resistor that goes from the output of the transformer to a previous stage is GNF. The transformer taps that go to the KT88 screen grids is LNF. One problem with GNF is that if overall phase information is changed by 180D the amp will squeal, perhaps when the tone controls are in a certain position? Just a guess.

Later still... Reviewing the Dyna Mark III, I see that the GNF goes from the output transformer to the cathode of the pentode. Generally in guitar amps the GNF goes to the tail resistor of the Phase Invertor (PI) leaving the tone controls out of the loop. I would for sure try removing GNF.

Also, what I remember from school 20+ years ago is concentrating all the gain in a single block (here the pentode) makes for a noisy circuit. Which, guessing again, is why guitar amps use distributed gain in the form of several gain stages.

Don

Ps:
http://www.aikenamps.com/
Read the technical stuff, mighty good, mmmm.

[Scott Anderson]

Shannon, I'm gathering the information. I'll post again when I have it and send you some pics. Can you receive a zip file?

Don, About the GNF and LNF difference. Thanks that was my guess. The noise issue is 2 fold. The bass I use and the fact my house is 50 years old. Much of the noise goes away when I'm connected to a proper 3 wired plug.

I'm thinking maybe I have a wiring error. Wouldn't be the first time. I was have trouble getting enough bias voltage only to fine that I used a 10K resistor instead of a 1K resistor.

The thing that gets me is that I followed the Sunn schematic fairly closely as I could. The only differences were that I'm using the 10 ohm cathode to ground for KT88s. I couldn't find a 12pf cap, pin 6 of the 6AN8, so I used a 10pf. If the was a success back in its hay-day, why am I having the problems. Unless I have a wiring error

[Shannon Parks]

Shannon, I'm gathering the information. I'll post again when I have it and send you some pics. Can you receive a zip file?

Hi Scott,

That'll be fine.

If the was a success back in its hay-day, why am I having the problems.

That's the good thing about starting with a real production schematic - you <know> you should have success. We'll figure this out.

Shannon

[Scott Anderson]

Shannon,

I've emailed you some pictures. Here are my voltage measurements by tube and pin number. EH Gold 12AX7, 1-220v, 3-1.69v, 6-223v and pin 8-1.67v. RCA 6AN8A, 1-276v, 2-117v, 3-127v, 6-117v, 7-30.4v, 9-.73v. EH KT88, 3-507v, 4-506v, 5- -60v, and pin 8-1.4v. The bias is set at -61.4v. Power supply voltages are, A-415v, B-410v, C-337v GZ34 pin 8-523v. I let the amp run for about 30 minutes before I took the measurements. There was still some drift , more noticably for the KT88s about 2 volts at pins 3 & 4.

As far as the squeal. I can adjust out the squeal but here we go. With the bass and treble min and both boost switches open, just turning up the volume cause squeal. If I set the bass max, close both boost switches, I can adjust less than 1/4 turn on the treble before squeal.

Don, Just an FYI. I jumpered out my bias resistor and didn't notice much of a difference. I guess maybe a switch to take out the bias resistor using it when you what to measure bias and taking it out when you don't. Just a thought. I will mess around with the GNF a bit to see what it does.

Whan I adjust the squeal out of the amp it sounds good but better when I bypass the preamp and use my SansAmp direct box.

[dhuebert]

When you remove the GNF, does the squealing stop?

Don

[Shannon Parks]

First off, check out this cool tool from Duncan Munro:
http://duncanamps.com/tsc/download.html
The bass and treble section from the Sunn looks a little like the Vox one.

The Sunn's first section with the volume control is a bit screwy, IMHO. The frequency response flattens out as you crank up the volume - no wonder some amps sound better cranked. OK, at a 12 o'clock position, we can assume a 50k/450k setting on the volume pot (as it is a log pot). The .002uF and the 50k to ground are a -3dB of around 1600Hz and is causing your thin sound, even with the boost enabled. You would need to crank it to 'ten' to eliminate this. For now, why not remove the .002uF cap and play with the amp a little bit. We may still need to fix the squeal issue, but we should be making huge improvement. Checkout that software, too!

Shannon

[Shannon Parks]

Here is what is going on in the volume stage. I suggest bringing up a pair at a time and toggling bewteen them from the taskbar.

Spice pic of first stage with 100mV source

With the Volume at 12 o'clock (50k - 450k)
With the Volume at 12 o'clock with Lo Boost (50k - 450k)

With the Volume at '8' (166k - 333k)
With the Volume at '8' with Lo Boost (166k - 333k)

With the Volume at '10' (500k - 0)
With the Volume at '10' with Lo Boost (500k - 0)

OK. I will go out on a limb and say the settings at '10' are more like what we want. Basically flat with the 'lo boost' enabled, otherwise about a -3dB dip. We could remove everything and use a much bigger coupling cap and a smaller audio taper (maybe 100k or 250k). I'll spice that later.

Shannon