DIY audio: what are you building?

[mwhouston]

From the data I looked at, too me, I thought the E88CC a high gain tube. Bruce of oddwatt agreed the E88CC a higher gain tube than the one he designed the preamp for. I compared the tubes at the same HT. The gain varied with HT. 

[unclekanus]

1 hour ago, Red MacKay said:

The 6DJ8 is quite unique as it is a very high transconductance TWIN triode.  e80cc is probably the closest as a twin triode.  It's actually a very nice tube with a bigger envelope and more current power.  Many use it to drive output tubes.  Gold pins etc, Not cheap as a result.  From memory, gain is similiar to a 12AU7.

 

Hey Red,

 

I found the Tungsram e80cc sounded better than the Euro version. It's also much cheaper! Yep, can replace a 12au7.

 

Also ran across a single triode that looks very interesting and is similar the the 6c45, called 6s4p. A "DR" version is available and is a 10,000 hour rated tube and it has gold grid wire. Can be had for about $6 a tube which is a heap cheaper than 6c45's these days.

[aussievintage]

16 minutes ago, mwhouston said:

From the data I looked at, too me, I thought the E88CC a high gain tube. Bruce of oddwatt agreed the E88CC a higher gain tube than the one he designed the preamp for. I compared the tubes at the same HT. The gain varied with HT. 

Nearly every sheet I have seen so far quotes 33 at  100 volts.  What is the voltage across them in your circuit?

[mwhouston]

230V

[aussievintage]

Just now, mwhouston said:

230V

Across each valve, anode to cathode?  Sorry I thought this was an SRPP circuit - you'd need a very high HT I would think.

[mwhouston]

You might be right. I’ll check the suggested HT. 

[aussievintage]

2 minutes ago, mwhouston said:

You might be right. I’ll check the suggested HT. 

You've got me doubting myself now too    I assumed the figures in the datasheets were anode voltages.

[aussievintage]

When I see

 

 

I thought that 91V is what is actually across the tube.

Edited April 26, 2020 by aussievintage

[andyr]

I'm building an AKSA 'Paris' (JFET-based) head amp for @MattyW.

 

This is a very interesting build because Matt wanted it to be able to cope with 3 situations:

1. accept 2 inputs - as he has 2 arms on his TT - and select one or the other to go through to the head amp PCB.
2. but be able to bypass the head amp PCB, for when he mounts a 'normal' HO MM/MI on one of his arms.
3. and have all switches - plus input RCA sockets for load plugs - on the front panel.

 

Matt has a large collection of carts which span:

• normal HO MMs/MIs
• rare LO MMs/MIs - such as a Pickering XLS-4500S (0.33mV), Stanton 785LZE (0.6mV) and Grado 'Statement' series (0.5mV)
• and LO MCs.

 

He has been using various SUTs with his LOMCs but SUTs are no use with his LO MMs/MIs as they deliver a completely wrong loading to the cart.  The Pickering and the Stanton like to see 100K, whilst the Grado (if it's anything like a 'normal' (HO) Grado) will probably like 33-35K.

 

With the default load resistor on the head amp set to 100K, the load which the cart sees can be adjusted by using load plugs (which work in parallel with the default 100K on the PCB).  This arrangement works equally well for his LO MMs/MIs ... and his LO MCs. 

 

Here are some pics:

 

The actually head amp PCB (small!!  64mm x 58mm).

 

 

 

The front panel:

 

 

 

The back panel:

 

 

 

Andy

 

[mwhouston]

Bruce’s suggestion is 215 on the plate of the active load. I’m about 230. Easy enough drop with a 5W limiting resistor. There is a 4uf PIO final HT cap in the preamp. A further resistor before the cap will add further filtering not that it is needed.  The HT is choked filtered with 400uf in a double (soon to be a triple) Pi. There is DC on the heaters. 

[aussievintage]

22 minutes ago, mwhouston said:

Bruce’s suggestion is 215 on the plate of the active load. I’m about 230. Easy enough drop with a 5W limiting resistor. There is a 4uf PIO final HT cap in the preamp. A further resistor before the cap will add further filtering not that it is needed.  The HT is choked filtered with 400uf in a double (soon to be a triple) Pi. There is DC on the heaters. 

So then, subtract the cathode voltage and is it about 100V ?

[mwhouston]

2 minutes ago, aussievintage said:

So then, subtract the cathode voltage and is it about 100V ?

Agreed about 100V across each tube. . 

[aussievintage]

Just now, mwhouston said:

Agreed about 100V across each tube. . 

 

So, sorry, but that makes the gain about 33 as I was saying.

[MattyW]

I'm very much looking forward to the Paris head amp.  I'll finally be able to try my low output moving iron carts and see what all the fuss is about. In theory it should better moving coil.

[aussievintage]

4 minutes ago, MattyW said:

I'm very much looking forward to the Paris head amp.  I'll finally be able to try my low output moving iron carts and see what all the fuss is about. In theory it should better moving coil.

I'd be interested in how well it works too.

[DQ828]

20 hours ago, Red MacKay said:

 

What are the 12AT7s doing in your circuit?  What gain is required from them?  Can you give up a little gain?

 

 They are the output valves in this circuit, the Broskie Unbalancer.

 

John put me on to a guy who had built several of these using the TP DAC and he was a wealth of information, and seem to do lots and lots of tests and comparisions so for the moment I'm following what he recommended, he seemed like the eternal tinkerer where I'm more a set and forget guy.

 

I tried cleaning the pins but dont think it made much/any difference.

 

If I'm going to measure Vk for the output valve I'm wondering where I should measure, at the top of R13 & 14?

 

Johon say "In the Unbalancer circuit, the differential input stage sees a constant-currentsource at its coupled cathodes, which allows the circuit to even better ignore common-mode signals and automatically sets the desired cathode-bias voltage for input triodes."

 

With that  in mind where should I mearuse Vk on the input valves?

 

I had a bit more of a listen last night and all in all I'm thinking the preamp sounds pretty good except for the underlying noise. Part of my problem is, Im always listening to my main syatem in the dead of night and where I am is very quite so any noise is easy to hear. One side has more crackle that the other so I'll do some valve swapping to findout which valve is causing the issue.

 

 

[mwhouston]

Definitely spoiled for choice at the mo. The one in play, in my main rig, is “Continuum” with the E88CC tubes. Just dropped HT down from 230V to 215 and added another cap to the heater DC. Sound very good. Slowly getting some hours up in the new tubes and parts. 

[DQ828]

Very nice

[andyr]

 

Another project I'm doing during 'lockdown' is a 40w Class A amp designed by Hugh Dean and described on DIYAudio, here:

https://www.diyaudio.com/forums/solid-state/344540-alpha-nirvana-39w-8ohm-class-amp.html

 

It's 40w into 8 ohms ... with +/-27v DC rails; I'm building a pair of the 4 ohm version, as I intend to use 2 stereo amps with my Maggie mid panels & ribbons (mids, 3.2 ohms and ribbons 2 ohms).  So I guess they will output 25w or so, into 8 ohms - which will be plenty for the 2 ohm ribbons, anyway! 

 

 

 

 

At the front is a power supply for each channel (each has a 300VA toroid).  These PSs are designed to be able to deliver 5amps continuously, with low ripple; see here:

https://www.diyaudio.com/forums/group-buys/336479-slb-smooth-butter-active-rect-crc-cap-mx-class-power-supply-gb.html

 

The cool thing about their design - and that of the amp boards (mounted on the rear panel) - is the devices that need heatsinking (2x transistors in the case of the PS, 2x MOSFETs in the case of the amp) can be remotely sited, connected using Molex 4-pin connectors.  The PS transistors actually don't need that much heatsinking (I over-subscribed!!  ) while the amp's MOSFETs are clamped between the Dell finned CPU coolers and the 5m aluminium L-brackets they are mounted on.  Noctua 14mm thick, 92mm fans are mounted on the case bottom ... with 92m holes cut directly underneath the fins.  So this should keep the MOSFETs acceptably cool! 

 

Each pair of MOSFETs is mounted on a 'Snubber PCB' - which contains some SMD components, as well as the MOSFETs.  A closeup of the way they are mounted is shown here:

 

 

 

 

The amp PCBs themselves are mounted on the rear panel; underneath them are speaker protection PCBs - see here:

https://www.diyaudio.com/forums/group-buys/340694-ready-run-rtr-ssr-dc-speaker-protection-delay-gb.html

 

Now I have to wire up this first amp ... then do it all again for the second amp!    At least all the PCBs for this, are done.

 

Andy

 

 

[MattyW]

Damn, that looks awesome Andy. I've never seen CPU heatsinks used in Audio gear before.  Your build is causing me to salivate!

 

Combines my old hobby with audio.  I'd love to do with old Thermalright heatsinks as I could never afford them back in the day.

 

Due to the far higher efficiency of those heatsinks I daresay you won't even need fans?

 

How much does such a build cost. Purely out of curiosity.... Not that my curiosity doesn't often lead to purchases.

Edited April 29, 2020 by MattyW

[andyr]

59 minutes ago, MattyW said:

Damn, that looks awesome Andy. I've never seen CPU heatsinks used in Audio gear before.  Your build is causing me to salivate!

 

Combines my old hobby with audio.  I'd love to do with old Thermalright heatsinks as I could never afford them back in the day.

 

Due to the far higher efficiency of those heatsinks I daresay you won't even need fans?

 

Yes, possibly - it would be nice if it works out that way, Matt.    The ideal would be if I only needed to put the fans on in summer - so we'll have to see.

 

Each heatsink has to remove up to 120w of heat.  (I say 'up to' as, strangely enough, Class A amps send less heat into their heatsinks the louder they are played! 

 

59 minutes ago, MattyW said:

How much does such a build cost. Purely out of curiosity.... Not that my curiosity doesn't often lead to purchases.

 

It's not a cheap build.  I had anticipated it might cost me around $3K to build 2x stereo amps.  In fact the total bill will be around $4,700, for them both!    The biggest single expense was the cases - around $1,100 for two (including all the hole drilling / engraving on the front & back panels.).

 

Andy

 

[MattyW]

120W is less than the old AMD Bulldozers used to output so I think you may have a chance of no fans.

 

I dare say the end result will be worth it. I may sell off a whole bunch of stuff so I can trade in my LTA MZ3 for a Z10 integrated.  Hopefully the tape out can be adapted for preamp use. It would really simplify my power on/off procedure.

Edited April 29, 2020 by MattyW

[mwhouston]

2 hours ago, andyr said:

 

Another project I'm doing during 'lockdown' is a 40w Class A amp designed by Hugh Dean and described on DIYAudio, here:

https://www.diyaudio.com/forums/solid-state/344540-alpha-nirvana-39w-8ohm-class-amp.html

 

It's 40w into 8 ohms ... with +/-27v DC rails; I'm building a pair of the 4 ohm version, as I intend to use 2 stereo amps with my Maggie mid panels & ribbons (mids, 3.2 ohms and ribbons 2 ohms).  So I guess they will output 25w or so, into 8 ohms - which will be plenty for the 2 ohm ribbons, anyway! 

 

 

 

 

At the front is a power supply for each channel (each has a 300VA toroid).  These PSs are designed to be able to deliver 5amps continuously, with low ripple; see here:

https://www.diyaudio.com/forums/group-buys/336479-slb-smooth-butter-active-rect-crc-cap-mx-class-power-supply-gb.html

 

The cool thing about their design - and that of the amp boards (mounted on the rear panel) - is the devices that need heatsinking (2x transistors in the case of the PS, 2x MOSFETs in the case of the amp) can be remotely sited, connected using Molex 4-pin connectors.  The PS transistors actually don't need that much heatsinking (I over-subscribed!!  ) while the amp's MOSFETs are clamped between the Dell finned CPU coolers and the 5m aluminium L-brackets they are mounted on.  Noctua 14mm thick, 92mm fans are mounted on the case bottom ... with 92m holes cut directly underneath the fins.  So this should keep the MOSFETs acceptably cool! 

 

Each pair of MOSFETs is mounted on a 'Snubber PCB' - which contains some SMD components, as well as the MOSFETs.  A closeup of the way they are mounted is shown here:

 

 

 

 

The amp PCBs themselves are mounted on the rear panel; underneath them are speaker protection PCBs - see here:

https://www.diyaudio.com/forums/group-buys/340694-ready-run-rtr-ssr-dc-speaker-protection-delay-gb.html

 

Now I have to wire up this first amp ... then do it all again for the second amp!    At least all the PCBs for this, are done.

 

Andy

 

 

Wow that’s extensive. 

[Gee Emm]

5 hours ago, andyr said:

 

   The biggest single expense was the cases - around $1,100 for two (including all the hole drilling / engraving on the front & back panels.).

 

 

That is crazy money for 2 chassis

[aussievintage]

9 minutes ago, Gee Emm said:

That is crazy money for 2 chassis

Glad I wasn't the only one thinking it.  Hey it's DIY, make them yourself