Phono Transfer System Construction Information

Construction information for DIY projects, including the MS Mid Side Matrix, Elliptic Equalizer, Mastering Console, Phono Transfer System, Insert Switcher and the Dual Class-A Amplifier. You can post your baby pictures here.
Gold
Posts: 677
Joined: Sat Dec 01, 2007 5:20 pm

Re: Phono Transfer System

Post by Gold »

I had a couple reasons for settling on the DB15 instead of a DB9. The first reason is that the 4pr. cable from base station to remote boxes fits better in a DB15. Reason number two is many lathes have SME3012 tone arms. You can see the ratty SME connector in the photos. A DB15 is damn close to a drop in replacement on an SME 3012 or 309. I converted my SME3012 to a DB15.
KMN
Posts: 53
Joined: Sat Jan 20, 2018 2:49 pm

My nearly completed build

Post by KMN »

All the little bits finally came together enough to allow me to throw this together. I'm no machinist but the electronics seem to be coming working as expected. The xlr connectors did not end up being orthogonal with the rest of the back panel though.

My original intention was as a fixed gain, balanced/differential path, dc coupled, record path vinyl ripping preamp with the record path taken right off the flat preamp board. The riaa monitor out is a nice bonus feature.

I had assumed I'd use it as a drop in my existing ripping equipment where where it would replace my riaa preamp. The chain currently includes a turntable -> riaa preamp-> mic preamps ->adc.

The build went smooth except I believe I had a short in the flat preamp pcb from positive rail to ground which I was able to clear with my benchtop supply.

I noticed that some of my That 1240's have sub millivolt offset per my meter while others have up to almost 6mV. I ended up sorting and ordered 6 more from Etsy store so I can sort for more low offset parts. right now I have 3 of the four on the flat pcb which have sub millivolt and one with 3.2mV offset. I'm hoping the next batch of six has one more sub millivolt offset part to complete my set. It's being nitpicky sure but it isn't going to hurt anything have extreme low offset since I am planning to try a dc coupled record path.

It was pretty intense cutting the heatsink down to size on my table saw so it would fit inside the chassis. I thought I might die.

I don't have a signal generator in my lab at home capable of reliably putting out 5mV so I wasn't ablree to go through all the test stages yet. I will probably have to try using my calibration vinyl and output from my turntable to finish the setup and testing

weird things I did:

- I used panel mount 5 pin xlr's for input and flat out and have a trs connector for the single ended output.
- I did the two wire wall wart thing hoping there would be fair level of transformer balance and with the hope there would be a noise advantage to this approach. note that I'm not doing too much that should cause an imbalance as I have no relays, etc.
- i did a series bead - capacitor to chassis ground, ssort of LC emi filters right at the xlr i/o pins. slightly visible in the photo.
-i ended up stuffing the optional input that1240s on my monitor stage (riaa eq switcher pcb) telling myself the result would place less of an impedance imbalance on the output of the flat phono stage, ie my record out signals.
- in my system shield gets earthed at the turntable end much like the diagram Wayne has posted. i utilize a single point ground and attempt to avoid impedance sharing in return current paths and avoid current loops wherever possible.

Recently I've had second thoughts about a couple things.
1. the reason I haven't wired my TRS in the photos and finished stuffing the montor board is I changed my mind and I'm changing over to a balanced differential monitor output and ditching the single ended/trs approach. I still need to order another 5 pin xlr panel mount.
2. I saw the discussion about remoting the pots on the flat board and how it can be done without jeopardizing common mode rejection by utilizing the cap minimizing shielding technique. on the potentiometer wiring. This intrigues me in that I could potentially eliminate my mic preamp stages if I could get this working for me. There is plenty of room on my front panel for a couple pots. The problem I foresee is I'm not convinced the That 1240's on the output of the flat preamp pcb would be a great choice to try and directly drive my adc's with even if they are probably perfectly capable of driving my mic preamp ribbon inputs. The adc inputs undoubtedly like a bit more current to be driven well. I could see trying to kludge in a couple that1606 off the flat phono instrumentation amp outputs to drive adc's with. Ideas?

I think it turned out kind of nice looking. Cost enough for the parts. haha.
Attachments
20180120_134619.jpg
20180120_125922.jpg
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Phono Transfer System

Post by mediatechnology »

Very nice Kevin!

Sawing that heatsink on the table saw was definitely an adventure. You made it all fit!

The one thing you did that I would change is shield the twisted pair lines for the phono inputs. Event though its in an enclosed box I think it will be quieter.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: My nearly completed build

Post by JR. »

KMN wrote: Sun Jan 21, 2018 4:45 am All the little bits finally came together enough to allow me to throw this together. I'm no machinist but the electronics seem to be coming working as expected. The xlr connectors did not end up being orthogonal with the rest of the back panel though.

My original intention was as a fixed gain, balanced/differential path, dc coupled, record path vinyl ripping preamp with the record path taken right off the flat preamp board. The riaa monitor out is a nice bonus feature.
I do not expect much benefit from DC response in a RIAA playback system... vinyl can exhibit some funny business down below audio frequencies.
I had assumed I'd use it as a drop in my existing ripping equipment where where it would replace my riaa preamp. The chain currently includes a turntable -> riaa preamp-> mic preamps ->adc.

The build went smooth except I believe I had a short in the flat preamp pcb from positive rail to ground which I was able to clear with my benchtop supply.

I noticed that some of my That 1240's have sub millivolt offset per my meter while others have up to almost 6mV. I ended up sorting and ordered 6 more from Etsy store so I can sort for more low offset parts. right now I have 3 of the four on the flat pcb which have sub millivolt and one with 3.2mV offset. I'm hoping the next batch of six has one more sub millivolt offset part to complete my set. It's being nitpicky sure but it isn't going to hurt anything have extreme low offset since I am planning to try a dc coupled record path.
you can probably trim or servo out DC... I am not the local 1240 expert here.
It was pretty intense cutting the heatsink down to size on my table saw so it would fit inside the chassis. I thought I might die.

I don't have a signal generator in my lab at home capable of reliably putting out 5mV so I wasn't ablree to go through all the test stages yet. I will probably have to try using my calibration vinyl and output from my turntable to finish the setup and testing
classic way to deal with this is a resistive pad that can scrub off 40 dB or 60 dB so your signal generator can happily make mV signals. You can choose R values that deliver a light load to the signal generator and nominal source impedance for typical cart (1.5k? + some inductance). Back in the day when I was doing a lot of RIAA work I had an inverse RIAA pad to make checking out RIAA paths easier.

have fun

JR
weird things I did:

- I used panel mount 5 pin xlr's for input and flat out and have a trs connector for the single ended output.
- I did the two wire wall wart thing hoping there would be fair level of transformer balance and with the hope there would be a noise advantage to this approach. note that I'm not doing too much that should cause an imbalance as I have no relays, etc.
- i did a series bead - capacitor to chassis ground, ssort of LC emi filters right at the xlr i/o pins. slightly visible in the photo.
-i ended up stuffing the optional input that1240s on my monitor stage (riaa eq switcher pcb) telling myself the result would place less of an impedance imbalance on the output of the flat phono stage, ie my record out signals.
- in my system shield gets earthed at the turntable end much like the diagram Wayne has posted. i utilize a single point ground and attempt to avoid impedance sharing in return current paths and avoid current loops wherever possible.

Recently I've had second thoughts about a couple things.
1. the reason I haven't wired my TRS in the photos and finished stuffing the montor board is I changed my mind and I'm changing over to a balanced differential monitor output and ditching the single ended/trs approach. I still need to order another 5 pin xlr panel mount.
2. I saw the discussion about remoting the pots on the flat board and how it can be done without jeopardizing common mode rejection by utilizing the cap minimizing shielding technique. on the potentiometer wiring. This intrigues me in that I could potentially eliminate my mic preamp stages if I could get this working for me. There is plenty of room on my front panel for a couple pots. The problem I foresee is I'm not convinced the That 1240's on the output of the flat preamp pcb would be a great choice to try and directly drive my adc's with even if they are probably perfectly capable of driving my mic preamp ribbon inputs. The adc inputs undoubtedly like a bit more current to be driven well. I could see trying to kludge in a couple that1606 off the flat phono instrumentation amp outputs to drive adc's with. Ideas?

I think it turned out kind of nice looking. Cost enough for the parts. haha.
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Phono Transfer System

Post by mediatechnology »

The sole reason for DC-coupling the low impedance outputs of the flat preamp is to eliminate electrolytics in the potentially-long balanced low-impedance lines from the remote preamp to the EQ box. Maintaining common mode rejection to DC in the balanced preamp stages eliminates the need for matched capacitors. The following flat path stage in the RIAA EQ board, after common mode rejection is realized, has a film HP coupling cap.

Phono preamps don't need infrasonic response.

I would take the flat record out from the balanced flat output of the EQ board for the reason JR stated. Its HP-filterd by a film. (C33,C34).

In my comparison using analog RIAA EQ vs. DSP in the good plug-ins (Olaf's and Wayne Stegel's) the only difference you see is that the DSP ones had gain to DC. The analog, when nulled to DSP, had deep nulls above 100-200 Hz. Naturally, below 100 Hz, the DSP ones accurately followed the RIAA curve. What that told me was their filters were as accurate as my analog ones and both were accurate compared to an I-RIAA Jung/Lipschitz network.

Despite their accuracy below 100 Hz I do think the HP response available in the analog path from the film cap is really needed during signal capture due to the infrasonic crud.

If DC and infrasonic crud makes it into the converter it should be HP filtered in the DSP domain to keep it from using up bits.

If DC doesn't make it to the converter it's likely already HP-filtered at the A/D driver by an electrolytic in order to strip off the converter's Vocm.

Servos are HP filters and they sometimes excel making a small-value capacitor look like a big one. But if you already have a high-quality film in the path why emulate it?

So better to HPF with a high-quality film early on. Thus, C33 and C34 and the big honkin' films.

DC offset at the output of the INA is usually very low with both bipolar and FET op amps if the cart is wired fully-balanced. So most of the additional offset are the 1240s. If you want to cherry pick those go for it but I'd still AC-couple it in the EQ board.
KMN
Posts: 53
Joined: Sat Jan 20, 2018 2:49 pm

Re: Phono Transfer System

Post by KMN »

I've been telling myself the sights I'll see by not HPF everything in analog may not be pretty and it's probably mostly an irrational goal I'll soon regret after trying dc coupled.

I've also been fascinated to compare the record path in the riaa eq switcher with a full balanced approach and see what is possible re hum rejection in my system.

My more rational side has already been telling me pretty much what you guys have said.I just haven't overcome the desire to go see these things for myself.

As Wayne and I have been chatting a little about on another board, I've really wanted to have a platform to compare some different infra sonic filtering approaches in analog and dsp.

I've also always wondered what the real world trade offs in common mode rejection I potentially give up by converting to single ended stages in the circuit. I have no doubt the result is well below the audibility threshold especially in high quality circuits like the riaa eq switcher. The trimmed that1240's have some amazing specs in that regard. But is the difference measurable in my system? Best case, I've previously reduced hum to somewhere around -100dB in my ripping rig using unbalanced stages where convenient and that was hard for me to accomplish with the windows pc DAW in the setup. I have yet to get my hands on a really nice quiet full balances stage to play around with.

I also have an experimental circuit of my own I'm working on which is doing weird things like dc servoing and and balanced/differential signal paths and I'd eventually like some hardware to compare against and these boards seem perfect.

I'm just really stoked about the platform I now have which can make all my experiments a reality.I don't mind reinventing the wheel if that is what happens. For me it's just a hobby and seems like fun.

Really great kits Wayne. I'll probably have to order your mid/side, elliptical filter circuits soon and make myself an analog vertical removal tool. BTW. I have made nearly zero progress on the related dsp tool. I have some of my own boards to release which have to come first and I never get time to work on my own stuff. Sorry about all the delay. I still want to finish that, just not yet.

Best regards and thanks everybody for the really great input,
Kevin
KMN
Posts: 53
Joined: Sat Jan 20, 2018 2:49 pm

Re: Phono Transfer System

Post by KMN »

I will pad the signal from my benchtop signal generator and upgrade my input wiring to STP.

I believe my RME ADI-2 pro ADC and Millenia HV-35P mic preamps both offer a direct coupled mode but am not certain of it on the RME.

Eventually I'd hoped to try the PCM4222 boards which are capable of dc mode if I'm not mistaken.

If everything came together right I could see the possibility of selling off all that expensive audiophile gear.

Even if I end up not caring about fully balanced and/or DC coupled paths at least I may not necessarily have to give up the flexibility of having those modes of operation available in case I ever change my mind.
Last edited by KMN on Sun Jan 21, 2018 3:26 pm, edited 1 time in total.
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Phono Transfer System

Post by mediatechnology »

KMN wrote: Sun Jan 21, 2018 1:09 pm I've been telling myself the sights I'll see by not HPF everything in analog may not be pretty and it's probably mostly an irrational goal I'll soon regret after trying dc coupled.
I'm usually the most freaked out about the LF wiggles when I'm looking at X-Y on the 'scope and I see LF Side vectors. When you have a lot of that due to warp there's almost a Doppler shift/watery effect that goes away when the Elliptic EQ is applied. Better-sounding on 'phones too. I'm just starting to really experiment with that.
KMN wrote:I've also been fascinated to compare the record path in the riaa eq switcher with a full balanced approach and see what is possible re hum rejection in my system.
I have a client that bought two RIAA-EQ/Monitor boards just to do that very thing. IIRC he built one of them to check it out single-ended and I don't know if he went on to build the second to fully-balance it.

I can't speak from experience going fully-balanced on the EQ boards. The source of hum in that board can be, like any other, radiated into the EQ's film caps. That's the entry vector I found and it goes away once the board is shielded.

To me the real benefit from balanced operation is the actual cart interface breaking away from co-axial RCA connections and going to fully twisted pair. Most carts are wired floating at least to the arm base. Some have twisted pair up to the RCA. You can't really do anything about magnetic hum appearing in the cart windings differentially but the common mode hum in the cable is more than people think. That's the magic.
KMN wrote:As Wayne and I have been chatting a little about on another board, I've really wanted to have a platform to compare some different infra sonic filtering approaches in analog and dsp.
I think you'll find the EEQ along with some gentle Left/Right HPF work pretty good.
KMN wrote:I've also always wondered what the real world trade offs in common mode rejection I potentially give up by converting to single ended stages in the circuit. I have no doubt the result is well below the audibility threshold especially in high quality circuits like the riaa eq switcher. The trimmed that1240's have some amazing specs in that regard. But is the difference measurable in my system? Best case, I've previously reduced hum to somewhere around -100dB in my ripping rig using unbalanced stages where convenient and that was hard for me to accomplish with the windows pc DAW in the setup. I have yet to get my hands on a really nice quiet full balances stage to play around with.
Since your boards are all in one box I would just run a single 1240 on the flat preamp and bypass the 1240 on the EEQ. Go directly into the gain trim stage. I don't think you'll see a measurable difference maybe a tad quieter on electronic noise floor. The balanced line between the preamp and EQ switcher was to drive long tie lines from lathe to transfer console.

Balanced EQ stages seem like an opportunity to double errors from component tolerance.
KMN wrote:My more rational side has already been telling me pretty much what you guys have said.I just haven't overcome the desire to go see these things for myself.

I'm just really stoked about the platform I now have which can make all my experiments a reality.I don't mind reinventing the wheel if that is what happens. For me it's just a hobby and seems like fun.

Really great kits Wayne.
Thank you! I think you've captured the essence of the RIAA EQ/Switcher. I put two signal paths, a bazilion jumpers and options into that thing so people could experiment.
KMN
Posts: 53
Joined: Sat Jan 20, 2018 2:49 pm

Re: Phono Transfer System

Post by KMN »

I think your point about how the signal is more susceptible to picking up common mode noise down at the cart is a great one. In what's hopefully a relatively quiet enclosure when the conversion to single ended is done up at the riaa eq switcher pcb, there is far less ability for the signal to pick up common mode noise. That seems certain.

I also agree with your point about balanced eq and component tolerance. Though, how good could it be if you planned to hand trim the caps and resistors for your one off?

These all sound like things I may not be able to resist experimenting with.
KMN
Posts: 53
Joined: Sat Jan 20, 2018 2:49 pm

Re: Phono Transfer System

Post by KMN »

I think you may be reading my mind.
Post Reply