6x 6550/kt88 bass amp build

[black_labb]

i've posted questions and posted references to my plans for this build so i thought i'd post some progress pics and a bit of a description.

there are 3 parts of the chassis'. one for the preamp, one for the transformers, output tubes and PI and the last one for the power supply caps and some controls. the second 2 would have been a single one had i been able to find one large enough, but i couldnt.

there are 2 torroidial power transformers. one for the plate supply, and one for the grids and the preamp. the output transformer is a hammond 1650W, a very big and heavy output transformer, should be up to the job.

the preamp has a 12ax7 and a 6bl8 triode/pentode tube. the controls are gain 1, gain2, tight, bass, shift (shifts the adjustment points of the james tone stack) treble and volume. the 2 channels are parrallel triode gain stages with slightly different operating conditions. they can be bridged by the switch below them. there are also controls on the bottom chassis which are a presence and focus control. they just alter the feedback loop. the presence control is something i thought of that allows reducing and increasing the feedback for the high frequencies, as opposed to just reducing. i'm not sure how well it will work out, but i will see. there is also a balance DI output that is drawn from the output of the amp which is what the XLR connector is for. the pot next to it is the level control.

the phase inverter uses a 12ax7 followed by cathode follower for each side of the signal from a 12at7.

the output tubes are set up fairly traditionally. there are bias settings for each pair of output tubes and there are bias points that are available to be tested on the rear of the amp (will be hidden behind the rear panel incase someone likes playing with screwdrivers). there are also fuses on the cathodes of each pair of output tubes. this should allow the amp to keep running when a tube goes bad instead of blowing the main fuse and causing alot of confusion. there is also a fuse for the secondaries (tied to a shared ground for the plate and grids supply), a fuse for the grid's/preamp supply and of course one for the mains connection.


the head case is going to be covered with black vinyl/tolex type material, but i will have timber front and rear panels. there will be cutouts for the controls and possibly ventilation on the front panel, and the rear panel will allow ventillation and will have 2 small fans attached to it. the timber for the visible panels is victorian blackwood (australian timber). the pieces can be seen in the images.




the above photo shows essentially how the chassis' fit together. the preamp chassis is connected to the front panel around 15mm higher than it currently is. the output transformer has its feet pointed down (i put it upside down so i didnt rest 13kg on the leads coming out)



this is the rear view of the bits together. i needed to have the filament connections come out of the preamp chassis anyway so i dicided to have each one go into the chassis just next to the tubes to avoid having them near much of the circuiry. guess its a form of shielding ignore the many leads floating around from the transformer and such..



this shows the blackwood panel that will be on the front and the rear of the amp.



this shows the circuitry thus far. the preamp is done aside from connecting the ground, filaments and B+ leads. the power supply caps are mounted on some boards i got at an electronic shop for power supplies which work quite well to mount them and for their circuitry. the power board is around 3/4 finished. the PI is finished (but i need to move the input jack), the output tubes are mostly finished (just need to connect the cathodes to the bias test points and set those up). i need to do the rectifiers and the bias supply and connect the transformer leads to the circuit. the output transformer requires more complicated impedance switching controls so i am using a switch and 2 separate connections to allow 4, 8, 16 and i think 24 ohm impedance taps. this still needs to be wired up.

[Steve Conner]

wee, what a monster!

I hope those bolts and brackets don't form a shorted turn through your toroids, otherwise you'll mostly be playing "Great Balls Of Fire".

[black_labb]

Originally Posted by Steve Conner

wee, what a monster!

I hope those bolts and brackets don't form a shorted turn through your toroids, otherwise you'll mostly be playing "Great Balls Of Fire".

yeah, quite a monster. i'll have to weigh it when its done.

the brackets are connected at one side only for that reason. i dont want that version of great balls of fire being played through this amp.

[black_labb]

had another late night and the wiring is done aside from the DI.
the powerboard seems to be wired correctly. had to flip the plate connections to the OT to stop the feedback from being positive and giving a nice howl. i'm just testing with a pair of 6550's as opposed to the full 6.

the preamp works in that i can get signal (only out of one channel, but i always seem to have issues wiring input jacks, input 2 is grounded somehow, will fix). there doesnt seem to be enough gain though. i think there is a problem with the circuit as opposed to the design.
the first gain stage is a 12ax7. the gain stage runs into a 1M volume pot followed by the pentode on the 6bl8. there is a 270k resistor after the volume control to work as a mixer (the other channel runs parrallel to this one in the same way but different setup of the triode). the 12ax7 isnt driving the pentode enought to produce any clipping or similar, even with a hot bass signal.
the supply voltage is at 280v with a 100k plate resistor and diodes on the cathode (thought i would experement a bit) the bias voltage is at 1.23V. the plate resistor is dropping the voltage on the plate to 172v.
i'll fix the second input jack and see if that gain stage is doing what i want it to do.
if anyone has any insights please share. thanks

[andrew_k]

Faaaar out, it's a monster!

[black_labb]

Originally Posted by andrew_k

Faaaar out, it's a monster!

well thank you

i'll get more pics up in the next 24 hrs

[black_labb]

these photos were taken shortly after connecting the 2 sections of the bottom chassis together and mounting the transformers. i have moved the smaller torroidial over 10mm since the photos as it was sticking out a few mm too far over the side of the chassis.



heres the front of the power board. it is mounted with some different 6550's that i have around (some pulled from other amps)



the rear



the view from above.





these 2 shots show the amp inside the boards that will be made into the head box. the chassis has since beem able to be centred after moving that torroid over a bit.



here is the circuitry. i will clean up the plate wires (i switched them over and left alot of extra wire there incase the oscillation was not from the feedback loop). i will also do what i can with the section under the 2 torroids, but when you have 24 wires coming in from transformers its a bit hard to not have them all in the same spot.

there is a spare 6.3vac winding there that i plan on using for a 12vfan (voltage doubler into a 12v regulator). i am tossing up wether i should mount the fans on the top of the amp or onto the rear panel. the top seems like a good idea except it means that the amp will have open vents on top of it so that spilled liquids can get onto the tubes. shouldnt be too much of an issue as long as i am careful.

[black_labb]

got a signal generator today. will have a look on the scope to see whats going on in the next week.
i'm pretty busy until after the weekend

[black_labb]

had a bit of a play with the signal generator and scope. was fun to see the waveforms, but didnt help. i took out the tight control and when i finished that i noticed that there was a missing ground which meant that the cap from the screen grid to ground wasn't grounded. this was why it was staying clean. i connected that up and the pentode is now overdriving.

i am going to reconnect the tight control which is currently bypassed. i have been done some playing with values for the pentode and am happy with the sound.

when the gain at full or close to the sound is as expected until i play 2 separate notes at the same time (bass guitar, guitar may be ok due to less low frequencies). when 2 separate notes are played there is a scratchy cutting out type noise. would this be blocking distortion?
unfortunately i cant test it out with a scope unless i can make 2 separate signals with the signal generator. i guess i'll have to try a few things out. the screen bypass cap is larger than i would have otherwise used, so i may try dropping the value to a more common value and see what effect the tight control has on this sound/effect

[defaced]

unfortunately i cant test it out with a scope unless i can make 2 separate signals with the signal generator.

I don't know if this will help or not, but it's the first thing that came to mind. If you can take two tones and merge them, you'll effectively get what your amp sees coming from your bass. It looks like this software can do that, but I've never used, it was just one of the first things to pop up on google when I searched for "audio generator". This plus your soundcard might get you what you're after.

Tone Generator Software - Audio Test Tone Generator Download

[Steve Conner]

Yes, it's probably blocking distortion. If you're using a pentode, the higher gain available in one stage may make matters worse.

If it only happens when two notes are played together, it could also be thought of as intermodulation distortion. A 2-tone test is the standard way of measuring IMD, and the easiest way of arranging a 2-tone test signal is just to get two signal generators and connect them together. One of them could be your 50Hz mains through a transformer.

To get good sounding distortion on bass guitar, you often have to filter out the low end before the overdriven stage, and then boost it back up again afterwards.

[black_labb]

thanks steve and defaced. i will reconnect the tight control which should help the distortion sound. it was put in place so that i could try to get some decent distortion and some decent clean on one channel. i will do some tweaking and see how it ends up.

[black_labb]

reconnected the tight control and spent the afternoon reducing cap sizes and playing around while thinking that it wasnt doing a thing. now i realise the pot must be faulty but measuring good (had it happen before where a small bit of solder or similar was connecting the wiper to the case, breaking it with a screwdriver fixed the pot even if it always measured as good) i'll pull the pot out and try another tomorrow.