D3a Phonostage reloaded Assistance required

What people are working on at the moment
Post Reply
User avatar
thomas
Old Hand
Posts: 910
Joined: Sat Sep 26, 2009 1:52 am
Location: N.W.Kent

#1 D3a Phonostage reloaded Assistance required

Post by thomas »

OK Its been cold and wet outside so I've been thinking….

I've dusted off last year's homebrew phonostage build and had a go redesigning it. Only this time I thought I might try and do this more properly...

For the first stage I'm going to leave as triode for now (if I pentoded it where do I put the feedback?!), but I thought a second stage D3a or e810f or Ruski equivalent would give it more gain. I've also replaced the mu follower end, which I was never altogether sure I'd optimised, for a cathode follower, which I know even less about! I tried a dc'ed ecc88 at first but I wasn't happy with my back-of-envelope calculations (I thought there was questionable linearity at the 160v ish working voltage unless I upped the current, but that resulted in a stupidly small Rk on the CCS section…). So I've gone for a 5687, with active load. I haven't actually built the CF bit yet, but the rest of the riaa is pretty well all there, tested, and reasonably 'flat' but I am going to change the Rl's from 7K2 to 12K, which apparently seems to halve the admittedly small h3 at the expense of slightly more h2; and led bias' for 1.2v IR types (from 1.8v red ones) on the D3a's (? grid current? and/or headroom questionable second stage??)

So…my first question is… does my CF section look reasonable/workable/ non lethal?!

Secondly when I thought I'd downloaded LTspice for mac I actually hadn't. I have been using this a bit http://www.trioda.com/tools/triode.html although some of it seems a little at variance to my eyeballing curves….

So, big ask, any chance of a sim (Izzy Wizzy Stephen I'm looking at you!) and any comments/ criticisms welcome.

Oh yes, here's the circuit...
Attachments
D3a reloaded
D3a reloaded
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#2 Re: D3a Phonostage reloaded Assistance required

Post by Mike H »

Bit confused by how the pentodes are connected?

Personally I might consider a CF after the 1st stage, will allow a much larger anode resistor and hence more gain.
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
thomas
Old Hand
Posts: 910
Joined: Sat Sep 26, 2009 1:52 am
Location: N.W.Kent

#3 Re: D3a Phonostage reloaded Assistance required

Post by thomas »

Ah yes my slack drawing. I've always trioded D3a's by connecting g2 and g3 to the anode after reading something on the 'net about connecting them this way, I think it was a Mr Thorsten's idea. Dangerous place the net….
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#4 Re: D3a Phonostage reloaded Assistance required

Post by Mike H »

Ah yeah but no but, the middle grid is the screen grid, not the signal grid. (?)
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
thomas
Old Hand
Posts: 910
Joined: Sat Sep 26, 2009 1:52 am
Location: N.W.Kent

#5 Re: D3a Phonostage reloaded Assistance required

Post by thomas »

Ooops! :oops:

Well I know what I meant!

Will amend presently!

I could put a CF after the first stage. Not sure what you meant by increasing the anode resistor tbh. I'm not sure gain is going to be a problem here! And wouldn't I need another CF at the output?
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#6 Re: D3a Phonostage reloaded Assistance required

Post by IslandPink »

I haven't looked carefully at the 5687 bit, but the gain stages & RIAA look very similar to my own phono which I'll post tomorrow, having just drawn it out properly. I think you'll be safe at -1.2 or -1.3V bias on the D3a's, my own has values like that , at least on the second stage ( which in mine is grid battery bias ) . Parts choice is deadly important. Coupling cap ( O.1uF ) I would suggest REL PCU or Mundorf ZN and not much else ( I could give you a list of caps to avoid ! ) .

What's the 150 R doing after the 5687 CF - should that be in the cathode circuit ? - otherwise it's just adding to the otherwise low Zout.
"Once you find out ... the Circumstances ; then you can go out"
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#7 Re: D3a Phonostage reloaded Assistance required

Post by IslandPink »

I assume the cap value in the first RIAA leg to ground ( 500Hz turn ) is 120nF and not 120uF :oops:
"Once you find out ... the Circumstances ; then you can go out"
User avatar
thomas
Old Hand
Posts: 910
Joined: Sat Sep 26, 2009 1:52 am
Location: N.W.Kent

#8 Re: D3a Phonostage reloaded Assistance required

Post by thomas »

Ooops and more oops…
Yes thats 120nf ahem...
TBH I drew the front half from memory and too quickly it seems….
Here we go
Attachments
D3a amended2.jpg
User avatar
thomas
Old Hand
Posts: 910
Joined: Sat Sep 26, 2009 1:52 am
Location: N.W.Kent

#9 Re: D3a Phonostage reloaded Assistance required

Post by thomas »

Thanks Mark good to know 1.2v bias is OK with a D3a.

The 150r is a 'cathode stopper' and yes it does add to the Zout…..
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#10 Re: D3a Phonostage reloaded Assistance required

Post by IslandPink »

OK, nice bit of tippex, although I suspect the brush is a bit big, because you couldn't use it on the 'u' of the 120uF !

Does the 5687 CF at 4ma have any more 'drive' than a D3a at 13ma ? ( which is about 2.5k-ohms Zout )
If not, the CF bit is not worth having. Lower output impedance is all very well, but if it doesn't have current to back it up, then it's not going to sound great .
Are you limited on the power supply current, because otherwise I'd like to see the 5687 at eg. 15mA .
I doubt the 5687 will need a cathode-stopper at 4ma - its normally only something you need to use on very touchy things like 6C45pi, 437A or E810F at higher currents.
"Once you find out ... the Circumstances ; then you can go out"
User avatar
thomas
Old Hand
Posts: 910
Joined: Sat Sep 26, 2009 1:52 am
Location: N.W.Kent

#11 Re: D3a Phonostage reloaded Assistance required

Post by thomas »

IslandPink wrote:OK, nice bit of tippex, although I suspect the brush is a bit big, because you couldn't use it on the 'u' of the 120uF !

Arf! went into daughter's bedroom for the tippex once, didn't want to venture in there a second time!

Yes the CCS bit at the bottom of the CF seemed to involve a bit of a juggling act between decent sized Rk load versus decent current. I guess a ss ccs would have been much easier. I must admit I cribbed from Morgan Jones who drew a schematic for an ecc88 CF/ CCS with a humble 2ma running through it, though he didn't for a moment suggest using it as was….

And the power supply should be capable of quite a bit more- I was going to use cascaded DN2540 wotsits and VR tubes for 300v as per a previous suggestion of yours, actually. They seem to give a more 'organic' sound than zeners….

Oh yes, the 0.1uf coupling caps in there at the moment are bog standard wima pp's….
User avatar
Nick
Site Admin
Posts: 15708
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#12 Re: D3a Phonostage reloaded Assistance required

Post by Nick »

Yep, the 5687 as it is is a bit redundant, the d3a will do as well. If you were going to go with a CF on the output (and no reason why not), I would be thinking more like 15ma and make life simpler with a depletion mosfet active load. But I would try it without the CF at first assuming you are driving 50k or so loads.

Mike: The restriction on the load resistor for the d3a is the available B+, No point using a d3a and then passing 2ma through it. If you need more gain, another CCS will help there.

After you have that working Thomas, I would try swapping the RIAA network for a lower impedance one and see how it compares.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
thomas
Old Hand
Posts: 910
Joined: Sat Sep 26, 2009 1:52 am
Location: N.W.Kent

#13 Re: D3a Phonostage reloaded Assistance required

Post by thomas »

Thanks people, looks like the low current CF will be getting dumped, but I'll bear in mind a 15ma CF 5687 with a ss CCS under it, as a future option. Yes it will be driving a 50k pot as it happens….
Well the front half is pretty well already built, and a bit of a rat's nest it is, tbh (I showed it to DTB recently who I think was terribly polite about it!) I might have this going in a day or two….
By low impedance, Nick, how low can I go? I'm guessing no lower than 3-4 times ra of the D3a for some reason, so say 8-10k?!
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#14 Re: D3a Phonostage reloaded Assistance required

Post by IslandPink »

(i) I'd agree with Thomas that it'd be pushing your luck to go much below 17.5K for the network with ~2.5k Zout ( D3a ) driving it. Could try 10K, not sure if it would be A LOT better than 17.5K ...
(ii) I'm not clear why, for the CF the lower 5687 section needs a honking great resistor underneath it - what's wrong with the way I used a 6N6P on this ? :
http://www.audio-talk.co.uk/phpBB3/view ... p&start=30
( which is an Aikido CF btw )
..and take the ouput from above the resistor ( ie. cathode of the upper section ) if you need even lower Zout. The way I used it there has the advantage you can play around with noise cancellation .
"Once you find out ... the Circumstances ; then you can go out"
User avatar
thomas
Old Hand
Posts: 910
Joined: Sat Sep 26, 2009 1:52 am
Location: N.W.Kent

#15 Re: D3a Phonostage reloaded Assistance required

Post by thomas »

Well as I hinted at earlier my knowledge of CF's is pretty sketchy…! ...but in Morgan Jones the honking great 62K resistor in the cathode of the ccs ecc88 gives a lovely low gradient, low distortion load line with 2ma going through it. When I tried that with more current my gradient was awfully steep….my triode sim said distortion was up, (although it was just the ccs part I guessed it was still important?) hence I reduced the current to increase the load….and then switched to the 5687, which seemed slightly better…..but then, I know nothing :)

Could go very low if I bunged that CF in after the first valve before the network….!
Post Reply