RIAA 1k5 LCR

What people are working on at the moment
User avatar
Nick
Site Admin
Posts: 15751
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#106

Post by Nick »

If I re-drew it like below, would that remove some of the ambiguity you were referring to?
Well, no, all it shows now, is that whilst the simple depiction of loops can be helpfull, in reality they are just helpfull constructs and not anywhere near what is actually going on.

lets say for instance, you look at the star grounding you have created by taking wires from the three elements to a common point. What would happen if you started to reduce the lenght of the center wire? Eventually it would become zero lengtt, and you would be back to convetion all buss earthing with a earth point in the center. At what length would the "star" grounding scheme fail?

I agree has you have said before, that RF eathing is more complex, and distance is important, but thats because what we are considering as a ideal wire, actually has inherent reactance. In the case you have mentioned, the only thing that would cause a difference between what you have drawn and the simple bus earth, is the presence of reactance in the wires.

All I am suggesting, is that for the idea of the loops to be taken onwards, you need to consider these parasitic effects to make the current paths make sense.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
Andrew
Eternally single
Posts: 4206
Joined: Thu May 24, 2007 2:18 pm

#107

Post by Andrew »


They need not share a common ground plane but they do need to share the same unambiguous 0V reference/AC signal ground. As such, they need to be close physically. IMHO, tradtional star style wiring does not provide a totally unambiguous reference
OK I can see where you're coming from, the fact the phono is close to 53db doesn't help, with the S&B at 1:20 I must be getting close to 80db. I hope to drop the S&B to 1:5 as soon as I am able. There's good reason for that wanting to try this.

This is all a bit of a suck it and see experiment I was hoping to see if I could get from cartridge to power valve with 4 gain stages, there's a sort of cathode follower to drive the 300B's grid too.

I'm going to get the scope out again this week and have a poke about. I hope to have something for the Owston meet, so don't want to take a big rebuild in case I don't get chance to finish.

I need to digest what you have said ready for another revision, I'm still not convinced I haven't got some heater noise breaking through....

thanks for the help.

-- Andrew
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#108

Post by izzy wizzy »

Well, no, all it shows now, is that whilst the simple depiction of loops can be helpfull, in reality they are just helpfull constructs and not anywhere near what is actually going on.
That's exactly what they are; helpful constructs. Helpful in a way so that we do see there is more to this and so using this as a spring board to understand better what is going on. If you want, we can discuss this more technically but IME, that can be alienating to begin with.
lets say for instance, you look at the star grounding you have created by taking wires from the three elements to a common point. What would happen if you started to reduce the lenght of the center wire? Eventually it would become zero lengtt, and you would be back to convetion all buss earthing with a earth point in the center. At what length would the "star" grounding scheme fail?

This is excellent. The idea was to show the drawing and then go through a set of what ifs. OK, reducing the center wire to say a pointon the LCR network reduces the combined output loop of the 1st stage and input loop of the 2nd stage. IOW, it removes one of the unambiguous legs somewhat and its propensity to pickup interference. It still is however star earthing still and not a bus bar. A busbar is a large gauge conductor that has some length. Even using this method causes many ambiguous situations but I'd need another drawing to explain why. However to answer your question, at what lenght would it fail? I don't really know but rather than use a busbar, why not try a grounding strip, a low Z strip rather than a busbar? I did all this in stages where I reduced a busbar to a point, no leads to it longer than the leads the component comes with. It's too much like hard work hence the very small strip idea.
I agree has you have said before, that RF eathing is more complex, and distance is important, but thats because what we are considering as a ideal wire, actually has inherent reactance. In the case you have mentioned, the only thing that would cause a difference between what you have drawn and the simple bus earth, is the presence of reactance in the wires.
Quite so and the confusion these impedances have. Also many star schemes have long wires flying about in the air, no where near an earth plane and so pick up all sorts of rubbish. This is the major failing of many star schemes. Combine this with competing signals in the same wires and it all makes a difference.
All I am suggesting, is that for the idea of the loops to be taken onwards, you need to consider these parasitic effects to make the current paths make sense.
I completely agree. If you want, we can take it there next.

cheers,

Stephen
User avatar
Nick
Site Admin
Posts: 15751
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#109

Post by Nick »

I completely agree. If you want, we can take it there next.
That would be great, and hopefully of interest/use to people, but maybe you would like to create a new thread to leave Andrew to his LCR, and start with another copy of the circuit, then we can take it from there.

Its a good time, as I only had this conversation with Colin yesterday about phono earthing and how you had to consider the effect and current likely to pass through each wire.
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
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#110 Caps

Post by IslandPink »

Andrew
I have 4-off 470u/6.3 NXs . You could do the super-E-cap pairing thing and get c.1000u that way .
Shall I send them ( in a couple of days ) - if so , send address on a PM

Mark
"Once you find out ... the Circumstances ; then you can go out"
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#111

Post by izzy wizzy »

I'm still not convinced I haven't got some heater noise breaking through....
What makes you think that? I mean, what are the symptoms?

Do have anything here or online that would detail the heater arrangement you have?

cheers,

Stephen
Andrew
Eternally single
Posts: 4206
Joined: Thu May 24, 2007 2:18 pm

#112

Post by Andrew »

Pure guesswork at the moment, the plan is to get the scope out as soon as I get chance..just steadying myself for another run at it.....I'll let you know how I get on, thanks for the help.

-- Andrew
Andrew
Eternally single
Posts: 4206
Joined: Thu May 24, 2007 2:18 pm

#113

Post by Andrew »

Well I have to admit I've been enjoying the phono to much to get it back on the bench and look at the very slight bit of hum that remains., that will be a post Owston job now, I'm afraid.

Anyway, this one's going in a box!

Now, I originally envisaged the amp with a much higher B+ as it turns out I only use ~145v once regulated, before the reg things can be reasonably flexible. So I don't need the massive TX I have! In fact, it more of a pain in the posterior as I have to burn off the excess voltage. I do however, need plenty of current.

So, I'm thinking about a toroid and solid state diodes plus a delay and relay circuit to give the heaters time to warm up.

For example, Hammond do some reasonable high voltage toroids and there's also the place Colin got his from.

Toroids are also much easier to get in a box

What's the panel think?

cheers,

--- Andrew
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#114

Post by izzy wizzy »

Hi Andrew,

There's a bit of a trade off with toroids IME. On the comparisons I've made, I like the sound of EI lams better than toroids. The theory being, toroids have much better HF response and so let through more HF hash. They do have lower radiated fields compared to EI lams. I've always thought of toroids as a compromise (to get roundother issues) rather than a performance item though I bet Plitron would say otherwise :-)

cheers,

Stephen
Andrew
Eternally single
Posts: 4206
Joined: Thu May 24, 2007 2:18 pm

#115

Post by Andrew »

An in between could be an EI TX and diodes, it saves on a winding.

I was wondering aloud if I would hear the softer switching of a vacuum diode when compared to say modern soft recovery high speed types, especially on a regulated PSU.

cheers,

-- Andrew
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#116

Post by izzy wizzy »

I would say you probably will hear the difference but whether you think it worthwhile is another thing. Snubbering, which I have no experience of may help. Others report favourably.

cheers,

Stephen
Andrew
Eternally single
Posts: 4206
Joined: Thu May 24, 2007 2:18 pm

#117

Post by Andrew »

I need a final mains TX for the phono, so I am thinking of using a toroid so I can get it in a nice box without a transformer poking out the top.

The PSU is already solid state so a toroid and solid state diodes now
seem an appropriate choice.

I was thinking of using two toroids back to back, with 2 off 230v to 9v-0v-9v toroids from, for example, Rapid (thanks to Nick for this idea).

My sums say the following which I thought I'd put up on the BB to show
how its done, but also in the hope someone might check I'm doing
this correctly. I'd rather have a sanity check as the consequences
might be nasty.

My HT requirement before smoothing/regulation is 150mA @ roughly 220v,
a cap input supply either CLC or CRC can be assumed, the HT solid
state regs follow this.

By placing two off the shelf toroids back to back I hope to put myself
in the area where I can get mass produced toroids which are cost
effective.

The first toroid will step down 230v to 9v-0v-9v the heaters will come
off this using center tap full wave rectification.

The second toroid will be connected in reverse with the 9v-0v-9v
winding connected to the first 9v-0v-9v winding. This will step back
up to 220v or so, accounting for losses.

The current drawn on the second toroid's 'secondaries' is 150mA.

so, 0.15A * 240v = 36VA

The primaries of this toroid share the load across both 0v-9v windings
so that's in effect 18v of windings.

36VA across 18 volts on the 'primaries' is 2A per 'primary'.

Now if I add in the 18v * 1.5A I need for the heaters which are tapped at this point that gives me another 27VA

27VA + 36VA = 63VA now add 25% for safe keeping means the first toroid
needs to deliver 80VA.

The second toroid needs to deliver the 36VA * 25% = 45VA (say 50VA)

This is about £22 from Rapid + VAT etc.

Another options could be a custom toroid, but I'm not sure what this might cost. Any ideas?

cheers,

-- Andrew
Andrew
Eternally single
Posts: 4206
Joined: Thu May 24, 2007 2:18 pm

#118

Post by Andrew »

After hearing Will's set up I'm tempted to choke load the D3a, any have any suggestions along this line, good ones to try etc.

thanks,

-- Andrew
Will
Old Hand
Posts: 497
Joined: Mon Jun 11, 2007 10:35 pm
Location: Northumberland

#119

Post by Will »

Andrew

Ive emailed this chap in Honkers to see if he is still
making chokes.
http://cgi.ebay.ca/ws/eBayISAPI.dll?Vie ... 0224992658
The 4 in my phono are from him, 230Hys.
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#120

Post by izzy wizzy »

Hi Andrew,

I presume you're choke loading the D3a in triode mode. If so, I think you'll like it :-) I use a Lundahl choke on a 5687 on the phono 2nd stage and it works well.

I would try and keep the heaters away from the same transformer as the HT. Rectification noise makes it back to the transformer and then you will be putting a step up transformer after that which might also amplify the rectification noise. Toroids are great at passing HF stuff and so are regulators. For the sake of one more seperate tx for the heaters, it might be worth it.

cheers,

Stephen
Post Reply