Long process Eros build completed!

[pRC]

This phono stage is replacing an original 'beta test' Seduction from way back. The Seduction has been a mainstay in my system with only a few tweaks for 14 years (added C4S, upgraded the caps, added a shielding box over the signal area) - holy crap, that makes it the oldest survivor in the music room. But as the rest of the system evolved and everything had been replaced (sometimes by several generations), you start to get itchy. I wanted to hear what a better phono stage could do and with great experiences with Bottlehead projects and the reports online and direct from other Dixie Bottleheads the Eros was a no-brainer.

Current Vinyl Setup:
Solid cherry re-housed Lenco/Bogen TT with Schick 12-inch tonearm
Denon DL103R in aluminum body
Altec 4722 mic transformers as step-up trannies, potted in maple case inside metal jewelry box
Bottlehead Seduction phono stage
Not-so-matic 57 preamp

The system sounds pretty great but I did have some ‘site’ issues -

• EM field from some high-voltage power lines next door (no house is perfect!)
• Ground loops to chase down (TT in on the opposite end of the room from the amp and speakers!)

Because of the EM field in the music room and the low-voltage MC cartridge I’ve had to shield all my interconnects carefully - so this Eros would be built to fit inside a Faraday cage. And that caused a lot of cascading changes so that I could access the RCA connections and also disassemble the thing later (sure you can just solder it all up in layers as you go, but what happens if you want to go back in?)

I went with a larger chassis - using off-the-shelf 12 x 8 Hammond steel boxes. I couldn’t find a tall box for the base so I ended up stacking two 12 x 8 x 2 boxes together. These cases have an open bottom and a solid top so I cut out the top of both leaving a flange of metal in place to attach bolts. Bought a pretty handy threaded insert setter tool for an earlier machining project, so I used a lot of threaded bosses to avoid fishing out wayward nuts later…

With a bigger footprint I moved the power transformer an inch further away from the shunt reg and signal area, and mounted a Triad CX7 choke on top next door. Choke has a grounded metal cage, though I didn’t know whether it would be needed for noise.

I wanted front-mounted jacks and controls so that I wouldn’t have to fish everything through the top cage later. The jacks and switches are mounted in a machined ebony backer plate that fits through the top of the cabinet and pivots into position behind the steel front panel. Made an aluminum L bracket to attach the ebony plate so all the signal wiring could be soldered and left undisturbed during assembly/disassembly.

I’ve been using vintage microphone step-up transformers for years to mate the MC cartridges I prefer to the Seduction inputs - I’ve collected several good sets ( see: http://cognitivevent.com/av_stepup.html ). I picked one of my favorite pairs to install inside the case with the Eros, some RCA MI-12399 trannies with a weird looks-like-an-octal-but-its-really-nine-pins base. I had potted these with beeswax into a drilled block of hard maple years ago, so I cut down the block to fit the new case leaving the wiring and jacks in place, then bolted the block into one corner of the top cage. 

The power input block is mounted to the back of the case, along with a chassis ground post. Again, too much behaviorally conditioning forced me to consider disassembling anything I build, so I ran a euro-strip along the side of the case for all the AC connections for the power input, power switch and neon indicator, power transformer, and chassis grounds.

I purchased the kit in 2015 and started on the project right away, but then it sat with the chassis partially completed for a long time. A really long time. Then in mid-September I kicked it back into gear.

[pRC]

Project Timeline
Kit Purchased late 2015 (plus, ordered a few bespoke parts and tubes)

Case construction late 2015 - mid 2016

• Hammond stacked steel 12x8x2 cases with tops cut out
• Hammond standard 12x8 perforated steel cage (5.2 inches high)
• 12 gauge Copper chassis top plate with extra space between 12BH7 and power transformer mounting
• Ebony front mount for connections and switches
• Milled aluminum internal supports for ebony front mount
• Stainless riv-nuts for enclosure assembly
• Stainless steel hardware throughout
• Copper standoffs for top plate (air gap around chassis inside cage)
• 12 gauge Copper mass-loading plate on cage top
• Expanded metal cage (AKA desk organizer box) around Triad choke in PS
• Aluminum bottom plate
• Oiled walnut end caps/supports

Started wiring September 2016

• Install RCA step-up transformers potted into maple case with wenge front plate
• Shielded wiring
• Dueland RIAA filter and output capacitors
• Caddock resistors in a few spots

Changes from Stock Eros Build
RFI/EMI Shielding
Internal step-up trannies (RCA MI-12399 microphone transformers)
Input selector switch - MM/MC - 4 pole, 4 position, rotary
Front-mounted RCA connections
Front-mounted gold-plated signal ground post
Front-mounted power switch and neon indicator
Rear-mounted OEM IEC Power module
Rear-mounted chassis ground post (also safety ground for metal case)
Euro-terminals for main power connections (for disassembly later if needed)
Stainless button-head hardware with nylock nuts
Vintage Cinch ceramic 9-pin sockets with silver contacts
Vintage Collins tube shields

PS Changes
Upgraded capacitors on PS board
Replace 220uF 250V Nichicon caps with 220uF 400V Matsushita cap
Triad CX-7 choke replaces 270 ohm 5W R
Euro-terminals to attach choke leads
(tuned V with inline 12W Mills 20 ohm R before choke)

Shunt Reg Changes
Replace .1uF Matsushita film caps with .1uF Sprague 600V 716P ‘Orange Drops’
(Replace one 220 ohm Carbon Comp R broken in assembly)
18ga copper ground lines from signal ground bus to Reg and PS

Shielded signal wiring
MC Step-up trannies to euro-terminal attachment
(Euro-terminals used to allow disassembly and load R adjustments)
MC Step-up euro-terminals to selector switch
MM input connectors to selector switch
(4P rotary input switch - MM/MC)
Selector switch to normal attachment points
Output signal to front output connectors

Signal Changes
Silver solder - Kester Lead/Tin/Silver
Internal step-up trannies
4P rotary input switch - MM/MC
Euro-terminals for step-up connections (disassembly and load R adjustments)
Euro-terminal for signal ground connections at normal attachment
Replace 75k ohm 1/2W R (RIAA filter) with Caddock .75W R
Replace .01uF Matsushita film cap with Dueland .01uF 400V (RIAA cap, direct connection to signal path)
Replace 1.5uF Solen film cap (output coupling cap) with Dueland 1uF 400V
Replace 47k ohm 2W R (output load) with .75W Caddock + heatsinks (calculated .62W on R)
Replace 470k ohm unknown R (grounding for output RCA) with Dale film R

Nerd Assembly and Testing Notes
Triad CX-7 PS choke measured less than 270 ohms, so voltage was a bit high. Added a 20 ohm 12W Mills R ahead of the choke to get input voltage to spec.

Shunt Reg passed initial test on first power-up

When I got everything assembled I had some issues to debug:
Measurements were off from the manual (using my trusty, dusty Fluke 26 III meter)...
Checking in here I see that others also reported this, and most were getting the same readings I saw

C6 was supposed to measure 39K ohms, but was 155 ohms (turned out to be one of the Caddock 47K R substitutions - marked as 47k but it was really a 47ohm R in disguise)
...and then everything was working fine!

Then, two days later when I went into the music room for a vinyl fix one of the channels was not working. Back into the case I go -
Argh, top cage is too hard to remove with wired up trannies… had to disconnect the step-ups inside the lower case to get to the tubes (this is going to be an issue for tube-swapping!)

Dead channel de-bug -
C4S LEDs are all lit.
Hook up signal generator to MM inputs and measuring output level and frequency
Oops, outputs RCAs are backwards (L is R, R is L) -
AHA, nothing from Right channel of 6922.
Swap in a different tube - bingo, fixed.
Noticing that the output voltage was off between L and R…
Crossed RCAs - (head slap) output caps are connected wrong at T6 and T7 - fixed

Worried about output mismatching I start swapping EF86 tubes, not much effect on the outputs
The 6922 tube is a different story - about 12 on-hand - swapping and measuring until the channels are nice and close.
Button everything back up and test again - Yippeee!
Back into the system for an extended listening session… everything is excellent.


Top Cage Solution: Cut hole in the top under the copper vibration plate, add piano hinge to copper plate and threaded inserts to close everything up EM tight - this will take more than a week to do right with machining and re-painting. Ordering a duplicate cage to install before I modify the original. In the meantime, no tube-swapping!

[ALL212]

That thing is NUKE ready!!  Zombie apocalypse ready also. 

Nice armor!!!

[johnsonad]

Thank you for the detailed build thread! Very interesting to say the least!  How is it sounding compared to your modded Seduction?

[JamieMcC]

I enjoyed reading about your build thanks, you sure have been creative and gone to a lot of trouble to minimise you issues with interference so hope thats worked out well for you.

[pRC]

Well the Eros was singing beautifully all day yesterday - I tried a lot of Genres: Jazz, Vocals, Pop, Classic Rock, Classical (chamber music and orchestral), Blues...

I had the Seduction in the system for a long time and some of the initial things I'm hearing are the shock of the new.

But, Wow. Very happy. The Eros seems to have more bounce and dynamics, better, cleaner bass, and more inner detail than the Seduction. I expect a lot of that is the MUCH better power supply in the Eros, since it looks like Paul Joppa has mad skills at RIAA.

Especially good moments: Nina Simone 'Little Girl Blue', Eagles 'What Kind of Love Have You Got', Holst 'Mars', ...


Once I get a few weeks of listening in, I'll swap the Seduction back in and try to write a more complete comparison.

/pRC

[BNAL]

Wow, what a nice build!

With all the mods that I have been reading about recently, and this one is the most ambishes, I might have to chance the caps and some resistors in the near future.

[pRC]

As many of you know we have some ferocious EM (electro-magnetic interference) in my listening room due to a run of Duke Power high-voltage line running next to our house. The music room is terrific other than the EM - 23 x 14 with a vaulted ceiling. Equipment is split into two groups - the turntable and 'winter' 845 amps are behind the listening couch, and the azurahorn speakers, preamp, and 'summer' 2A3 amp are up front. This means I have signal wiring running along the length of the room, about 32 feet per run.

The previous solution to this was standard RCA interconnects with an isolation transformer at the destination end to remove hum. This actually worked by the way. And over the years I’ve collected a few different isolation transformers, because, like everything else in this hobby, they don’t sound the same.

An aside on RCA isolation trannies:

• Best sounding - Jensen Iso-Max CI-2RR - $195 - RCA in/out
• Most flexible, cheap solution - Rolls ‘Buzz Off’ - $38 - good enough for a computer setup, this unit has all sorts of input and output jacks (RCA jacks + mono ¼ inch jacks + stereo â…› inch jacks)
• Others tried - JK Audio ‘Pureformer’ $65, RDL AV-HK1 Hum Killer $90, On the shelf still - Bogen DIY

The newly unleashed Eros phono stage has enough output oomph to drive the lines to the other end of the room directly - so I started thinking about how to build a better set of shielded RCA cables. Awesome fellow bottlehead Lawrence Yeatts (Knoxville TN)  suggested I’d be better off with a balanced cable and transformers at both ends, and he even picked the trannies and sent links!

Regarding the long signal run, I would use transformers at each end and do it balanced with unshielded twisted pairs.

The Jensen JT-11P-1 or similar bridging line input transformers are excellent for the "load" end of the run if their secondaries are looking into 10K load or greater.
 
For the source end you can use a "plate to line" transformer and Jensen makes a very nice one, the JT-10K61-1M.
 
When doing long runs I believe balanced lines with unshielded twisted pairs is the best way. However even if you don’t run balanced lines having a transformer on the load end to give galvanic isolation breaking the inevitable ground loop and keeping out RFI is very helpful.
 
See these links below.
http://www.jensen-transformers.com/wp-content/uploads/2014/08/jt-11p-1.pdf
http://www.jensen-transformers.com/wp-content/uploads/2014/08/jt-10k61-1m.pdf



Off to the interwebs!

Input box - steel for EM Faraday cage - 6 x 6 x 3.5 inches

• BUD Industries JB-3945 Steel NEMA 1 Sheet Metal Junction Box with Lift-off Screw Cover, 6" Width x 6" Height x 3-1/2" Depth, Gray Finish
• Jensen 10k:600 10k61-1m transformers
  http://www.jensen-transformers.com/wp-content/uploads/2014/08/jt-10k61-1m.pdf
• RCA inputs (Two sets - one front, one rear)
• 4-pole rotary switch for input selection
• XLR (male) output jacks
• Binding post for grounds
• Toggle switch to lift signal ground from chassis (alt: Buddha Camille’s yin/yang diodes?)
  

Output box - load transformers are smaller! - Steel again - 6 x 4 x 3 inches

• BUD Industries JB-3952 Steel NEMA 1 Sheet Metal Junction Box with Lift-off Screw Cover, 4" Width x 6" Height x 3" Depth, Gray Finish
• Jensen Low-power 1:1 11p-1 transformers
  http://www.jensen-transformers.com/wp-content/uploads/2014/08/jt-11p-1.pdf
• XLR (female) input jacks (from input box)
• One set of alternate RCA inputs (rear)
• 4-pole rotary switch for input selection
• Set of RCA output jacks
• Separate 2-pole polarity switches for L+R RCA outputs
• Binding post for grounding
• 2-pole toggle switch to lift ground from chassis

Steel junction boxes are drilled with step drill, using a dremel with cutoff wheel to even up the inside holes. Paint is silver hammertone. End caps are walnut. Top plates are steel with 13 gauge copper plates joined with adhesive. Hardware is stainless steel. Most mounting holes are threaded to ease assembly. Input transformers (Jensen 10k61-1m) are mounted with coils at 90 degrees. Output trannies (Jensen 11p-1) are potted into a hard maple block with beeswax.

Assembly and testing was almost without drama - I had one channel that did not pass signal consistently… turned out that one of the toggle switches I used for polarity switching on the outputs was bad!

First day impressions - Music is much better than the old RCA cables that are now abandoned in place - both with the input transformer box, but also bypassing the box using some XLR to RCA adapters connected to the Eros outputs directly (and using the destination balanced transformers). The input trannies do eat a little signal level, but there is still plenty of gain in the system. Noise floor is lower and dynamics sound better, more natural... we listened way too late into the evening.

Sort of an overkill and expensive DIY cable solution, but it works for me!

[ALL212]

Now 'dats some cool stuff there!   

[Doc B.]

You could always do what Patrick Gleeson did at Different Fur Studio in San Francisco and line your walls with copper foil.

[pRC]

You could always do what Patrick Gleeson did at Different Fur Studio in San Francisco and line your walls with copper foil.

That looks pretty cool in my mind - You know, at one time I did research special paint with metal shielding!

http://www.lessemf.com/paint.html

Heh. Looks like a lot of trouble. And what about the windows? I guess we could convert the room into a cave...

Nah. Blithe has been very forgiving of the music room, but even she has limits.

I do plan to panel the walls someday, using green glue between 1/4 inch cherry panelling and the existing sheetrock should give some acoustic benefit. But the horns are pretty directional anyway so not a lot of reflection energy in contention, and lots of shelves with books and albums to diffuse residual sound. I did have typical room bass issues with big peaks and nulls, but adding multiple subwoofers (I have 3 now) plus the dual Open Baffle 15 inch woofers under the Azurahorns... spreads out the room nodes and integrates a lot better. And with all those bass drivers they are all loafing along at low power most of the time.

There's a tip in there somewhere - in my experience, adding multiple cheap $200-500 subs is better than a single $2k sub.

/pRC

[Doc B.]

Aluminum screens over the windows might work, depending upon the frequency of the interference.

[pRC]

Aluminum screens over the windows might work, depending upon the frequency of the interference.

Yah, and there's the rub - it's 60Hz EM from the power lines, and theory (as I read it anyway, please straighten me out if this is crap thinking) says that I need ferrous shielding for a faraday cage to shunt EM around the room. A big steel box.

Should have bought a bank vault instead of a house...

BTW, they had an awesome anechoic chamber at GTRI when I was working at Georgia Tech - fully RF/EM/Audio shielded, for testing equipment emissions. An extremely odd and disconcerting feeling if you were inside and they closed the door... very, very strange effect on the mind. Not a place you wanted to hang out. Later, one of our offices in Atlanta was in an old bank building where they had a huge decommissioned vault (still had the huge door, but the locks were removed), it was not exactly the same oddness as GTRI, but there was an echo of the feeling whenever we closed the door while we were working in that vault.

Hrm, I haven't thought about that in a long time...

/pRC

[2wo]

Get a big roll of copper foil and wrap those pesky high tension lines. Careful to just ground one end though

[pRC]

Get a big roll of copper foil and wrap those pesky high tension lines. Careful to just ground one end though

As long as I get to wear my big copper suit while I'm doing that - Duke Power, look out!

(Hey kids, this is a joke, don't try this at home)


We're actually a bit lucky - it's a really old line built in 1919. It's low and hidden by trees; we can't even see the closest lattice pole except in winter. Duke can't up the voltage without replacing the whole thing, and they already have parallel lines that carry much more power nearby. Surveys for where the property lines and easement lines are well-known, but they don't match the actual center-lines for the installation, and we're in an old in-town neighborhood with deep political connections. So last time they did tree trimming on our lot we had the head of Duke's NC lobbying organization walking around the yard trying to make sure we were happy (got his Italian loafers a bit scuffed up).

On the other hand, more copper never hurts anything!

/pRC