Snail Phono Pre

What people are working on at the moment
little eddy
Old Hand
Posts: 693
Joined: Sun Nov 09, 2008 2:06 pm
Location: Manchester

#181

Post by little eddy »

Well I've fitted a 100k pot in series with 221k resistor as R1 of the RIAA network. With a pair of 5751 input valves, I started with the pot a zero and this definitely got rid of the overly strong lower bass. I decided that I would wind the pot up until a point where the bass is considered overly strong. So far it is 247k and still tweaking.

Depending upon what Rout is of the cascaded CCS, I had estimated that R1 would be in the range 254 - 262k.

Is this scale of 'error' or AOT adjustment normal on a phono stage?

I estimate that the ECC83 will require around 21k less resistance so may add a switch to alter the filter frequency as appropriate. I would want to keep the wires as short as possible but with R1 beneath the mounting plate, it will be dificult to access once the enclosure is all bolted back together.
TD-125/RB250/MC25FL & 'Snail' phono, NAS/SBT with CS4398 DAC, 41MP pre & MoFo Power, still messing with OBs.
User avatar
Nick
Site Admin
Posts: 15705
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#182

Post by Nick »

When I am designing a phono from scratch, I start with a spice sim, and then measure when built. As long as you arrange that the load on the first valve is such that the valve ra is less by a 1:10 or more its possible to build a accurate stage without any need to tweek. The last couple I have built have all been within 0.5dB from 20 to 20, but you need to be able to measure to that accuracy to know that, I don't think its feasible to do it by ear. A inverse RIAA, scope and square wave source is a good simple way of getting it right.
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
shane
Social outcast
Posts: 3403
Joined: Sun Sep 16, 2007 12:09 pm
Location: Kept in a cool dry place.

#183

Post by shane »

Tounge-in-cheek, but how do you measure the accuracy of the inverse RIAA?
The world looks so different after learning science. For example, trees are made of air, primarily. When they are burned, they go back to air, and in their flaming heat is released the flaming heat of the Sun which was bound in to convert air into tree.
Andrew
Eternally single
Posts: 4206
Joined: Thu May 24, 2007 2:18 pm

#184

Post by Andrew »

You don't, not normally anyway, tho' you could, and it can be done easily enough.

Being a pedant I did, but the standard way is to build a known circuit with tight tolerance components, measuring each one as you go. The whole lot should then be a sum of its parts.

If you really want to, then you plot the end result against the ideal curve given by the maths, either on a spreadhseet or graph paper.

Andrew
Analogue, the lost world that lies between 0 and 1.
User avatar
Nick
Site Admin
Posts: 15705
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#185

Post by Nick »

I did it the other way, First check that the inv RIAA + phono gives a flat response. Then measure the phono on its own and check that it matches the correct curve. Therefore the inv RIAA is correct.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
little eddy
Old Hand
Posts: 693
Joined: Sun Nov 09, 2008 2:06 pm
Location: Manchester

#186

Post by little eddy »

Nick wrote:As long as you arrange that the load on the first valve is such that the valve ra is less by a 1:10 or more its possible to build a accurate stage without any need to tweek.
Very good point when you haven't built up an array of test equipment. Starting with the high Ra ECC83 doesn't help but guess its a balancing act given the necessary gain. Having parallel valves reduced Rout to 68 - 100k but with no cathode bypass caps I end up with a ratio of 1:4 at best.

If I bypass the cathode resistors then the valve Rout falls significantly to between 25 to 40k and with R1 rising to circa 290k, I would have an Rout/R1 ratio of to between 7 to 12,

I bought 4 100uF Oscon caps following a 5751 RIAA discussion on the WD forum. Is it worth the tonal influence of a cathode bypass cap to give me this reduced criticality of R1 or simply recommended because I have no test equipment? It has the added benefit that switching between ECC83 and 5751 is nowhere near as sensitive in terms of RIAA accuracy.
Nick wrote:I don't think its feasible to do it by ear.
Agree but at least I could hear when its not right.
TD-125/RB250/MC25FL & 'Snail' phono, NAS/SBT with CS4398 DAC, 41MP pre & MoFo Power, still messing with OBs.
little eddy
Old Hand
Posts: 693
Joined: Sun Nov 09, 2008 2:06 pm
Location: Manchester

#187

Post by little eddy »

So I have now fitted bypass caps on the cathode resistors and the level of low bass 'emphasis' has reduced somewhat. And using the 5751, I now have an R1/Rout >10 so it should sound pretty balanced using the theoretical RIAA values.

So R1C1=2187uS and with C2=6n9, R1tot=316k.

Over a range of combinations with Ra between 50 and 58k, and Rl between 300k and 1M, R1 should be in the range 289k8 and 294k6. So I set my pot to the average at around 292k but the bass was still slightly noticeable and I found myself listening to this rather than the music.

Is there anyway I am undercalculating the Thevenin resistance that the capacitor sees? I am assuming the two valves in parallel with the CCS load so with an Ra of 58k and assumed Rl of 300k, I get Rout to be 25.5k.

I am assuming that R2 (46k1) below C1 does not need to be taken into consideration when determining R1 because the equation in MJ does not mention R2 and this is calculated in isolation soley based on C1.

Maybe I am just hearing the bass because it is more prominent than I've heard before, or maybe on particular recordings bass is louder than I would have imagined.
TD-125/RB250/MC25FL & 'Snail' phono, NAS/SBT with CS4398 DAC, 41MP pre & MoFo Power, still messing with OBs.
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#188

Post by Mike H »

Yes there can be very large variations of bass between recordings! :D

That's the trouble with using the same one(s) all the time as a test.

By putting the bypass caps in you've most likely boosted the top-end & mid-range more, hence the bass seems less. Well it is really.

Ideally you want to measure the phono pre output from a signal generator input, gain should be 10 times more at 20 Hz than at 1 kHz. And again 10 times less at 20 kHz than at 1 kHz.

Also ensure that the power supply isn't resonating (step response effects) to enhance the low frequencies.
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
little eddy
Old Hand
Posts: 693
Joined: Sun Nov 09, 2008 2:06 pm
Location: Manchester

#189

Post by little eddy »

Well I replaced the pot with fixed resistors in the RIAA with a switch to reduce it by 6k8 so I can use both ECC83 and 5751 input valves. With Mullard ECC83 things sound a lot more balanced so very happy.

Also fitted the chinese switched attenuator. Checked all positions and all seems OK. Definitely no worse than the Alps. Bass seems more tuneful and better separation. I now just need to remove the Alps from the input of my poweramplifier.
Attachments
Switched attenuator.jpg
TD-125/RB250/MC25FL & 'Snail' phono, NAS/SBT with CS4398 DAC, 41MP pre & MoFo Power, still messing with OBs.
little eddy
Old Hand
Posts: 693
Joined: Sun Nov 09, 2008 2:06 pm
Location: Manchester

#190 Re: Snail Phono Pre

Post by little eddy »

Can't believe it's so long since I considered changes to my Snail phono/pre, but now that I have my new 41MP pre-amp, time to review the phono section.

1st part is to increase my supply voltage by 20V allowing me to increase the currents, particularly the 6072 section at 2.5mA per triode.

Could I simplify the schematic and have a few thoughts:
1. Reduce the ECC83 grid stopper to 100R (as per I think Nick did on his WAD Phono)
2. Remove the 6072a grid stopper
3. What is the 1k phono series output resistor doing? I see this is the same arrangement as the WAD Phono but wondering can I remove and what is the risk?
Attachments
12-12-07 Snail 2-stage with single RIAA CCS-2 small.jpg
Last edited by little eddy on Tue Jun 11, 2019 5:26 pm, edited 1 time in total.
TD-125/RB250/MC25FL & 'Snail' phono, NAS/SBT with CS4398 DAC, 41MP pre & MoFo Power, still messing with OBs.
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#191 Re: Snail Phono Pre

Post by Mike H »

Good lord this goes back a while. :D

little eddy wrote: Tue Jun 11, 2019 3:32 pm 3. What is the 1k phono series output resistor doing? I see this is the same arrangement as the WAD Phono but wondering can I remove and what is the risk?
Not sure in this case, but in the transistor/op-amp world, it may be provided as some sort of short-circuit protection, if the o/p socket is shorted. Also, to prevent high-frequency reflections in a screened lead, but would have to be very high frequency, practically RF.
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
little eddy
Old Hand
Posts: 693
Joined: Sun Nov 09, 2008 2:06 pm
Location: Manchester

#192 Re: Snail Phono Pre

Post by little eddy »

I'm pretty sure I copied the output cap/resistor detail from the WAD Phono 2/3 design where they used an ecc83 as a cathode follower.

The Audio Note M7 phono employs a 6072 triode output and there is no series output resistor.

Edit - also the AN M7 6072 output triode doesn't use a grid stopper either.
TD-125/RB250/MC25FL & 'Snail' phono, NAS/SBT with CS4398 DAC, 41MP pre & MoFo Power, still messing with OBs.
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#193 Re: Snail Phono Pre

Post by Mike H »

little eddy wrote: Tue Jun 11, 2019 5:20 pm I'm pretty sure I copied the output cap/resistor detail from the WAD Phono 2/3 design where they used an ecc83 as a cathode follower.
OK, so just on the basis the designer can't know what's going to happen to the amp in service, so any kind of screened lead, short-circuit accidents, who knows. So like a safety thing. I can't remember using one like this, or at least recently, so I dare say you could omit it. I think what type of resistor it is might have an affect on what the signal sounds like too. My personal pref probably carbon comp or wirewound, IF I was going to do it.
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
little eddy
Old Hand
Posts: 693
Joined: Sun Nov 09, 2008 2:06 pm
Location: Manchester

#194 Re: Snail Phono Pre

Post by little eddy »

Based on a few of Nick's posts, I have ordered one of these inverse RIAA cards: https://www.ebay.co.uk/itm/173886602640?ul_noapp=true

I get how I can sweep through the frequency range using a sine wave input and measure the output of the phono to determine if I have a flat response.

But I have also heard of a 'square' wave test but trying to get my head around this. You input a square wave via the inverse riaa card and you should get a square wave out of the phono. If there is overshoot on the leading edge then do I understand correctly that this would suggest more gain for the higher octaves, and an 'overdamped' leading edge the opposite?

I then 'assume' that you can alter the series R1 to either increase or decrease the low frequency response of the riaa network to optimise either a flatter frequency response or a more accurate square wave output?

But what does this square wave test tell me that the sine wave frequency response wont? Or is it that with one square wave test and adjustment you don't need to do the full frequency response?
TD-125/RB250/MC25FL & 'Snail' phono, NAS/SBT with CS4398 DAC, 41MP pre & MoFo Power, still messing with OBs.
User avatar
Nick
Site Admin
Posts: 15705
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#195 Re: Snail Phono Pre

Post by Nick »

The advantage of the square wave is that you have all the frequencies of interest from the fundamental up in the one test, and to get a square wave out you need all the frequency and phase response to be correct. Its harder than you might think to measure a 1v RMS signal at 1kHz and 2kHz and know if they are within 0.1dB without proper test kit. A inverse RIAA, square wave source and scope will give you a good approximation to the same thing.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
Post Reply