Ultra-high performance low-buck phono preamp (DIY project!)

J!m

Veteran and General Yakker
Joined
Dec 24, 2019
Messages
12,152
Location
Connecticut
Tagline
BOT
So, about a year ago, I purchased a set of two boards to build both a MC and MM version phono preamp. But when I finally got around to doing something, I discovered a LOT had happened while I was away...

The design is by (Dr.) Wyn Palmer, who, in a former life, designed IC chips. So, this exercise was partly as a slap in the face of all the audiophools, when he decided to design and build a preamp that not only kicks serious ass of some serious preamplifiers out there (including some well regarded big buck ones by famous makers who will remain nameless D'agostino) but uses "normal" components and a CHEAP switching power supply!

Original link: (Version 1) High gain opamp based preamp for low output MC cartridges. | Audiokarma Home Audio Stereo Discussion Forums
(Version 2) Ultra high spec. opamp MC/MM phono, warp "elliptic" filter, line, headphone amps. | Audiokarma Home Audio Stereo Discussion Forums
Link on the builds: The Official "Build" thread for the WYN Phono,Line,Warp,Headphone Pre amp | Audiokarma Home Audio Stereo Discussion Forums

He also did a nice headphone amp I am considering for a later build. I suspect it will be much more resolving for my monitoring needs... (info on that in the Version 2 thread)

The highest attention was placed on getting the RIAA exact. And the deviation from ideal, is less that .3dB in any case as I recall (but I may be off on that), but suffice to say, the most accurate RIAA EQ as compared to several other well regarded (and expensive) phono preamplifiers that he tested. Several Audiophiles, with very expensive preamps, have sold them to be replaced by Wyn's design.

That's the good.

The "bad" news is, the board is (ideally) built to suit a cartridge. The loading and capacitance are adjustable (but generally fine left unpopulated) but the gain is set by a series of resistors in key locations in the circuit. They have to be carefully considered to retain the high accuracy of the RIAA. But, the frequency response is very flat from DC to ultrasonic. And, although it could be billed as "analytical", what you get is what you have. In other words, it doesn't add (or subtract) any "flowery" tube distortion, or enhance or soften any frequencies across the audible band (and beyond). What goes in, gets accurate RIAA correction, is amplified, and passed on down the line. (NOTE: One builder has done some heavy modification with a rotary selector for cartridge loading, and there are a host of other "mods" for this thing). Due to simplicity, the component selection is QUITE CRITICAL in some key places. Not expensive audio-grade components, or beer-can sized capacitors. Everything is carefully considered, and rational, based on science and testing (as well as , of course, listening!). Because of this, some of the components are often out of stock (I purchased from three vendors). But, in some cases tighter-tolerance components are available, and fine. One capacitor value is a 1% version and out of stock. But next week it'll be something else I'm sure...

For my build:

I am doing a two-box (power supply external), two-mode version. I will have two boards inside (plus the power smoothing board): one for MC and the other for MM. There will be four buttons on the front: MM/MC select, 10dB gain, Mono/stereo, and Elliptical Filter. The filter is a simple design that accounts for record "warp", where a very low frequency is transferred to the speakers. Sometimes you may see this as woofer excursion during the warp travel. Anyway, this filter simply eliminates this low frequency information, without any side effects at all. You loose zero music information. Its design is based on similar filters used on cutting lathes and in empirical testing, it works fantastically. A switch is really not needed at all, honestly, but it is switchable, so I'm switching it. If you saw my dumpster-dive records, you probably wouldn't bother with a switch either...

The component cost for my build is $365.24, plus ~$15 for the boards. A build box is needed for the preamp and (in my case) outboard power supply. (everything can go in one box provided the PS is shielded from the boards). I figure another $150 for build boxes. Under $550 for a world-beater. Building a single board would be much less obviously, as would eliminating some of the features the later board incorporates. And the external box is not needed either. This can be built quite inexpensively if it is a single-mode, dedicated unit in a single box (as originally designed). Believe it or not, the design does not get any traction on DIY audio, but, that's their loss!

If you like vinyl and building stuff, get on the band wagon! Wyn is a saint, and very helpful with technical problems too. It doesn't get any better!
 
Today I finally started doing something.

NOTICE: I need help with surface mount op amps. I have the adapter boards and chips but do not have the equipment, skill or patience to install them. Perhaps @Gepetto could help?

That aside, I began installing resistors and diodes (low stuff) so as not to waste too much time.
C569DF28-CA29-491F-957D-4AE02B54BB16.jpeg55E54B2A-5D51-4172-97DF-A97CB2FA30ED.jpeg
The second photo is the audiophile grade lead bender. Made of jet-age 6061 aluminum and cryogenically treated for grain alignment and lowest possible noise.

I have a good bit of the resistors installed as well as the diodes. I really need those SMCs done to go too much further.

And another member on AK designed a board for the four front panel switches. I ordered one (set of three, so I have two spares) and I received notice they shipped today.

I found a couple board labels that may be incorrect, so I have to check with Wyn on that. Otherwise it’s going fine so far (a lot of parts!)
 
SMDs suck. I hate dealing with those. There’s some kind of adhesive that you can use with them instead of solder, or so I’ve heard…
 
Low melt solder paste and ninja like speed and accuracy.

Or the correct equipment.

I have neither.
 
Where can i get this...audiophile grade lead bender. Made of jet-age 6061 aluminum and cryogenically treated for grain alignment
 
I believe Cardas sells them for a couple hundred. That other company that sells fuses for stupid money probably has their own version (gold plated) as well.

I got the board label sorted out- the value is correct but the number is wrong.

And I’ve been advised to “go for it” and try the SMDs myself…
 
I used to build SMD devices that utilize a computers soundcard and software for shortwave radio reception. I started out using solder paste and a hot plate, worked O.K.. Then tried hot air, worked O.K.. Afterwords I found I'd often go back in and touch up joints using magnification (Optivisor) and a long thin SMD tip on the Hakko soldering station.
For me, after all this, I found it easiest to SLIGHTLY wet a pad with solder, position the part, reheat that pad so the part just sticks. Then solder the other side, and finish that first pad. Good light, proper tip, good magnification, proper position, and two hands holding the solder pencil works for me.
 
Those are large SMD pads on that board. Just have the proper tip, small diameter solder and magnification before you go for it.
 
Last edited:
Looks to me like those designations will take a DIP device, or is there something else I'm no seeing?
 
When originally designed, the chips were DIP8 but that package is no longer available, so the adapter boards were made and the original board was modified to take both packages.

My fear is not precision (I repair watches) but rather overheating the devices. They are not super expensive but they are often out of stock.
 
Don't worry about overheating if your using the proper tip. The proper tip will be long and thin, low thermal mass, holds enough heat to solder SMD nicely, no good for soldering anything larger, the larger parts "suck" the heat right out of it. The trick to doing this is having the right tools, just like every other job. If you don't have the correct tip, stop, research, get it. 700 normal degrees Fahrenheit setting as normal.
It can be done with other tips, but results won't be as nice. If you try with a normal tip for thru hole components, make sure you have good solder wick available, you may need it.
 
Looks to me like those designations will take a DIP device, or is there something else I'm no seeing?
You are correct Lee, the board will take either. The device that goes in there may only be available in SMD however.
 
My experiment with SMDs was a disaster. An “Aurora” with a bazillion tri-color LEDs and an SMD resistor for each one. After this, I swore to never attempt another. It didn’t work, BTW…
8776DFCC-064F-4C6E-B318-C8DF2D916A3E.jpegA08417A3-9715-4B35-B46E-51844A055C1F.jpeg
Only the blue junctions fired…
C7963E7A-9115-4DA3-A28D-9DF1CA7F83D0.jpeg
 
Last edited:
Mark W., wow, that is a amazing amount of soldering. You did a really good job on that fine pitch IC on the LED side. Jim's pads are huge compared to it.
 
I bought some of those preamp circuit boards from the first batch. Then things started changing almost daily it seemed & I backed off for a bit & then forgot about it until now.
I have not been back to AK since they ruined the Bartertown.
Nice to see what you think Jim.
 
Today I finally started doing something.

NOTICE: I need help with surface mount op amps. I have the adapter boards and chips but do not have the equipment, skill or patience to install them. Perhaps @Gepetto could help?

That aside, I began installing resistors and diodes (low stuff) so as not to waste too much time.
View attachment 54919View attachment 54920
The second photo is the audiophile grade lead bender. Made of jet-age 6061 aluminum and cryogenically treated for grain alignment and lowest possible noise.

I have a good bit of the resistors installed as well as the diodes. I really need those SMCs done to go too much further.

And another member on AK designed a board for the four front panel switches. I ordered one (set of three, so I have two spares) and I received notice they shipped today.

I found a couple board labels that may be incorrect, so I have to check with Wyn on that. Otherwise it’s going fine so far (a lot of parts!)
What part number goes in the SMD locations?
 
Made a little progress today:

1BB17578-D92D-4334-907C-6BFF9FB8C74C.jpeg
I got the hang of the small chips after a bit. I installed the pins first and then held the “tick” in place with tweezers. Loaded the iron tip and brushed it across the part pad onto the board pad. It worked well until I hit a wall:

47C3BD4D-0A04-4ED1-B63E-A4FEED8D8394.jpeg
this one wasn’t happening.

The others aren’t pretty but they’re on and not shorted. Hopefully not more than medium rare.

57746EA3-2072-48D9-AF92-034C0045CA18.jpeg
 
Back
Top