Filters , poles, interaction, resonance

What people are working on at the moment
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#16 Re: Filters , poles, interaction, resonance

Post by IslandPink »

Now, I had never seen this - there are a lot of great things on Bjorn's site :
http://kolbrek.hoyttalerdesign.no/index ... r-supplies
Not read it all yet, either !
"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:

#17 Re: Filters , poles, interaction, resonance

Post by izzy wizzy »

Very interesting. The one that catches my eye is the compensated single VR tube; stable Z and phase over wide frequency spectrum. Food for thought.

cheers,

Stephen
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#18 Re: Filters , poles, interaction, resonance

Post by IslandPink »

That's what I use on the phono amp - 2-off OD3 in series. I haven't tested the compensation across them, I got approx. values from Gary Pimm, and it's around 0.1uF and 390R .
"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:

#19 Re: Filters , poles, interaction, resonance

Post by izzy wizzy »

I tried Bjorn's idea of approx 10u and 100R across a 0d3 to even out the response in my line stage. WOW! I did this months ago and haven't gone back for a comparison again.

By comparison, I would have said previously it sounded like the impedance curve in terms of expression but after it was far more even across the range. If that makes any sense ... so hard to explain. Much like hi Z EQ networks sound like the RIAA curve while low Z ones sound more even. Same kind of idea. It was nice to be able to thank Bjorn directly at Munich for his idea ... which he hasn't tried himself BTW :)

cheers,

Stephen
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#20 Re: Filters , poles, interaction, resonance

Post by IslandPink »

Nice .. I must try that myself then - so you went 100R and 10u across each OD3 - or do you only need 150V , not 300 ?
One must think the type of cap is important at 10uF - what did you use ?
I might just try going to something bigger than 0.1u but not quite so big , first , for a pair of OD3's.
"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:

#21 Re: Filters , poles, interaction, resonance

Post by izzy wizzy »

I only have one 0D3. If you look at the impedance curve, at low freq he measured the 0D3 at about 100R and with 10u and 100R (he used 91R and 9.1u), the tc is about 1kHz where the 0D3 starts to rise rapidly so from then on, the 100R starts to get to work in parallel and evens out the impedance of the shunt reg. With two in series, you might have to double the values R and halve the C to achieve similar. I didn't try anything else or take it out again which is something I should do for completeness.

OK now it gets embarrasing. I think I used whatever came to hand out of the cap box. IIRC, it's a WIMA MKC4 of about 35 year vintage which is a metalised polycarbonate but it's face down in the bottom of the box and it's very tight in there so can't say for sure.

cheers,

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

#22 Re: Filters , poles, interaction, resonance

Post by izzy wizzy »

Thought I'd better go revisit this 10u 100R bypass thing as quite a bit has changed. First I took it out and couldn't really tell. So I put a switch on it and can switch it live. Can't really tell any difference :oops:

This happened also on this line stage with the power supply cap. Standard BHC electro vs BG WKz. There was a difference but couldn't really decide. I wonder if this has something to do with it being differential? I dunno. Stupid hobby.

cheers,

Stephen
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#23 Re: Filters , poles, interaction, resonance

Post by IslandPink »

He he !
Well, I think it being differential will make the PS choices a lot less significant - it certainly seemed like that when I did the Aurora PP 300B amps. In the case of the phono you're not getting that close to full swing on the devices either, so the AC demand on the PS will be very small.
Must look at your last phono now - hadn't realised it was PP.
"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:

#24 Re: Filters , poles, interaction, resonance

Post by izzy wizzy »

Thinking back, I think the psu wasn't set up currents wise as well as it is now and maybe that's why the bypass made a difference; significant difference IIRC. It might also have been put in when I had the CCS in the tail without the tx on the input. Now it's so much better balanced, as you say, maybe it makes it less sensitive to the PSU.

And following on from that, I went back into the phono these last few weeks to see if a few tweaks might improve things. Had some better transformers to get the PSU performing more how I intended with good results. This allowed more current through the stages to good effect; all compromises I knew about and plthought not big, certainly now better. Tried dampers instead of the mercury and was very disappointed to find themercury so much better. Weird! I want to try mercury in the line stage and am hoping it makes no difference.

Bigger CC grid/screen stoppers on the pentode reduced noise so it's now completely inaudible with headphones off the line stage. We hear so many say pentodes are noisy and so shouldn't be used for phono. Real world would seem to indicate otherwise. Still got it a bit more microphonic than I would like.

I'm fiddling with the EQ to get better value caps in there which has been beneficial. It's quite easy to hear even though I'm only looking at 0.2dB deviation in the sim beforehand. We shouldn't be able to hear that but it's obvious.

It's got a PP output stage coz I had a PP tx and also I wanted to try DC coupling without massive decoupling caps. Seems to work but then I have no comparison.

cheers,

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

#25 Re: Filters , poles, interaction, resonance

Post by izzy wizzy »

Thought I'd post an update on my travels with glow tubes. I was using one in the line stage/headphone amp. I then tried doing the same in the phono output stage which is also a diff amp like the line stage. I thought it was better ... well it sounded more impressive so I thought, let's live with it a while. 2 months on and I'm not convinced. I was listening less and less and it was causing my ears to hurt. Kinda like an HF resonance that made my ears itch. I've returned to the cap decoupling and what a relief.

Then a day later I noticed something was odd, so I mono'd the output headphone connection and traced it to one channel in the phono. It sounded like the glow tube was back in but it wasn't. I then saw the decoupling cap had fallen off :oops: No cap was the same as the glow tube sound.

Then I modified the line stage from L C R 0D3 feeding both channels to LCLC then RC to each channel. This was much more how I like it and less of that glow tube sound.

Overall, I'm running the volume control 2 clicks lower than when all glow tubes were in for what seems to be the same impact/dynamics. And overall, I like the sound much better without the regs; fuller tone, better low level detail and far more relaxing.

I'm glad I did it but don't think I'll be using them again.

Cheers,

Stephen
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#26 Re: Filters , poles, interaction, resonance

Post by IslandPink »

Sorry a bit slow to reply Stephen, I saw it when I came back from hol and have been mulling over this for a few days. It's unsettling because I respect your ears but also it seems like the opposite result I had when I moved from a passive choke-smoothed supply to the VR shunt on the phono.
Can you just do a sketch of the circuit you used with the VR tube(s) with all details on there, just to be clear what you used ?
Doesn't have to be neat or anything, a wobbly 'Paint' drawing will suffice.
"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:

#27 Re: Filters , poles, interaction, resonance

Post by izzy wizzy »

I will do but for now, I too have mulled over what happened.

When I first did it, there was 1 less decoupling cap with the VR tube instead. I thought it sounded like less cap colouration and that was the positive. It seemed a bit tighter overall in a positive way. But over time it didn't do well for my ears.

And that might be where the difference is. A taste thing. I mean plenty of people rate regulation but every time I go near it, over time it just doesn't agree with me.

I find it interesting that the VR tube sounded like no decoupling cap when it fell off. There could be a clue in there. I might be trading cap colouration for VR tube colouration.

Cheers,

Stephen
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#28 Re: Filters , poles, interaction, resonance

Post by IslandPink »

I need to see the circuit. You seem to have used one VR tube, so that doen't match my expectation of a shunt for eg. 300V on the end of the power supply. In any use of that part, there definitely needs to be a zobel across one (or both) of the VR tubes, otherwise you'll get more noise from the tube, and the parasitic inductance which may affect the top end detrimentally.
"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:

#29 Re: Filters , poles, interaction, resonance

Post by izzy wizzy »

OK. I used one VR tube per channel on the phono 2nd stage which runs off 150V but no zobel so maybe that's why it went wrong. It's so easy to do, I'll have another go. Thanks.

I'll have another go at the line/headphone amp too. Something for the weekend :)

cheers,

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

#30 Re: Filters , poles, interaction, resonance

Post by izzy wizzy »

Hi Mark,

So I looked up RDH to get more info and couldn't find anything. Didn't have my reading glasses on so who knows. Likewise Morgan's book. Briefly looked on the net, nothing easily found. Looked at your circuit and you use 390R 0.1u across 2 x 0D3. Lynn Olson uses 1.5k 0.15u across 2 x 0D3 in one of the 2006 Karna ccts and 750 0.33u across 1 0D3 which is probably what I'll use for starters. Is this a suck it and see or is there some science in this?

Cheers,

Stephen
Post Reply