Nicks Aikido 300b amp

Andrew · Post by **Andrew** » Thu Jul 12, 2007 8:21 pm

Is that -105 db?

-- Andrew

Post by **Nick** » Thu Jul 12, 2007 8:37 pm

Yep, 24bit M-Audio soundcard.

Though, notice with the scaling used, the 1khz fundimental is at +15Db

Maybe I would reduce hum and noise if I put it in a box and used neat wiring methods

Darren · Post by **Darren** » Thu Jul 12, 2007 8:50 pm

Nick wrote:
Maybe I would reduce hum and noise if I put it in a box and used neat wiring methods

Doubt it !!!

Andrew · Post by **Andrew** » Thu Jul 12, 2007 8:52 pm

Looks impressive!

So 90db, am I correct? Still pretty, no very, good.

I'm getting to the point where tho' I would like new speakers, the 300B is actually driving my 89db floorstanders satisfactorily and with a little room to spare and without sounding too nasty.

I put on my CD player today, it been in the kitchen for a while, what a revelation, it made Metallica sound flat, lifeless and dull.

Also, I would hazard a guess that modern 83's don't actually manage a mu of 100 as DTB's 5751s are as loud as my modern 83's in the same spot.

I think now my driver PSU voltage is limiting the useful swing into the 300B. I wonder if I'll notice a 1uF cap before my choke input, probably not, especially when I stick in some 6N30 shunts I have wired up and ready to do, just short a filament TXs.

cheers,

-- Andrew

Post by **Nick** » Thu Jul 12, 2007 10:26 pm

So 90db, am I correct? Still pretty, no very, good.

The fundamental is at +15, the floor is at -10 (iah), so in theory 115dB, the reality may be a bit less, but there are no fudging of the numbers.

It will be a bit less if I include below 900hz. I will do that tommorow for comparison.

Andrew · Post by **Andrew** » Sat Jul 14, 2007 1:44 pm

Is that a Fostex box I can see in the picture?...How did the super twater work in the OBs?

cheers,

-- Andrew

Post by **Nick** » Sat Jul 14, 2007 2:20 pm

Still waiting for the super twater to arrive.

Wilmslow say they should have them in end of next week.

Andrew · Post by **Andrew** » Sat Jul 21, 2007 9:24 am

Hi Nick,

Do you bother to earth the secondary of the OPT in your 300B? I have a felling mine aren't but things seem quite happy. I guess I could get DC offset to ground due to capacitive coupling on the windings if they float. What am I missing?

cheers,

-- Andrew

Post by **Nick** » Sat Jul 21, 2007 9:37 am

No, in theory you should in case the insulation on the TX broke down, and you touched the speaker lead while holding on to a gas pipe.

Other than that I can't see it making a difference, other than increasing the PD across the transformer, so increasing the energy stored in the interwinding capacitance, but I guess thats a insignifican't factor.

But I spose, try it and see, is the answer, and let us kno

(Hmm, maybe I wil start posting in the style of N Molesworth)

I have all the bits for the 6c33c now (chokes arrived from Phil), so will create a thread for that later.

Andrew · Post by **Andrew** » Sat Jul 21, 2007 10:56 am

I can see why it would matter using feedback etc, but for a SET, there's no connections elsewhere, yes its safer, that's true, but otherwise, can't see anything else.

I've been thinking about your point about regulated supplies and Rp makes sense really as lower Rp more current to deliver so supply must have lower impedance for the same performance. Essentially, won't a lower Rp valve be harder to drive thus requiring a better supply?

Having said that, I suspect regulation will help with any valve tho', as Paul's EF1 PX25/PX4 into the 212 showed, I reckon a good bit of the sound he had was down to the PSU. Good bass....

Looking forward to the 6C33.

cheers,

-- Andrew

Post by **Nick** » Sat Jul 21, 2007 11:39 am

I've been thinking about your point about regulated supplies and Rp

Yes, that was my thought as well.

Another aspect I thought about was the lower Rp valves, use lower reflected impedance output transformers, so the stepdown ratio will be less as the Rp decreases, so as the stepdown reduces, so the effect of a given "movement" on the B+ (from either change in signal current against supply impedance, or ripple) on the signal the loudspeaker sees.

MJ gives (for a common cathode amplifier)

PSRR = (RL + ra) / ra

So for a 45 with a ra of 1k6 and a 5k load its 4.1, for a 300b with a 800R ra and 2k5 load, it comes out as 4.1 also

So our theory doesn't seem to work as I hoped

And for a 6c33c with a RL of 300 and a Ra of 80, its 4.75, interestingly more than the 300b. And if you use a 600R load as many do its even better at 8.5

However maybe if we then add the ratio of the stepdown. I am guessing that we can multiply the rejection ratio by the stepdown ratio to give the effective ratio

45 case 5K:8 = 25:1, so 4.1*25 = 102
300b case 2k5:8 = 18:1 so 4.1*18 = 73
6c33c case 300:8 = 6:1 so 4.75*6 = 28

And for a 6c33c with a RL of 300 and a Ra of 80, its 4.75, interestingly more than the 300b. And if you use a 600R load as many do its even better at 8.5

Of course there could well be an element in making the numbers fit the theory here, but what do folk think?

Andrew · Post by **Andrew** » Sat Jul 21, 2007 11:08 pm

I bit too tired to get a grip on this, I tried, I really did but after the PSU debugging tonight...my brains are mince....I'll try again to tomorrow, remind me as I think your onto something. I also suspect I was working on a conceptually dumber level as well, just that a lower Rp tube will simply just be harder to drive as it will draw more current. But then I also visualized it as a potential divider too, which I think you're doing as well. And to cap it all MJ bangs on about series shunt regs effectively being potential dividers but its one of those things you think you understood than something goes 'click' and a whole new layer of understanding opens up.

Have you factored in that its also easier to wind a TX for a lower Rp, less turns has to sound better; there's a 6C33C-B going on e-bay for $25 or so each, I'm tempted......no too much else to do....

cheers,

-- Andrew

Post by **Nick** » Sat Jul 21, 2007 11:18 pm

Have you factored in that its also easier to wind a TX for a lower Rp

Yes, thats another plus, the low Ra, means that primary inductance can be far far lower than normally needed, so again, less primary turns are needed, which should be a good thing, but then again, I know effectivly bugger all about magnetics, so maybe Paul has some views on the merits or not of output valves with a Ra of 80 ohm.

I expect I will find the problems that come with them once I get going. Still doing metalwork and woodwork at the moment, should be able to start the first one tomorrow with luck.

Of course the big downside is the curves show they are not the most linear of beasts, but I think the op point I am using will be good for 18w, so as I only need a watt or so, it may be ok.

Paul Barker · Post by **Paul Barker** » Sun Jul 22, 2007 7:50 am

yes as you would expect it's easier to get hifi bandwidth (a mythical goal that Frits never even though of, which far outreaches the abilites of our loudspeakers), but the overall sound of the amp will be dominated by the sound of the valve. The low output impedance will be seductive in bass, but unlikely the rest will sound as good as your 300b or 211.

Andrew · Post by **Andrew** » Sun Jul 22, 2007 9:31 am

Now there's a challenge

Why are you allowed to multiply by the winding ratio?

I was thinking of fig 5.26, as Rl reduces it gets harder to drive and the impedanc eof the supply needs to drop to maintain the same ratio.

But then I could be talking cobblers.

cheers,

-- Andrew

audio-talk

Nicks Aikido 300b amp

#31

#32

#33

#34

#35

#36

#37

#38

#39

#40

#41

#42

#43

#44

#45