Call Us: +1 720-406-8946

The CNC Phono Stage DIY | Page 2 | DIY Forum | Forums

A A A
Avatar

Please consider registering
guest

sp_LogInOut Log In sp_Registration Register sp_MemberList Members

Register | Lost password?
Advanced Search

— Forum Scope —




— Match —





— Forum Options —





Minimum search word length is 3 characters - maximum search word length is 84 characters

No permission to create posts
sp_Feed Topic RSS sp_TopicIcon
The CNC Phono Stage DIY
How to buy parts, build, and tune this cost-effective home made phono stage
Topic Rating: +1 Topic Rating: +1 (1 votes) 
March 9, 2017
7:01 am
Avatar
Admin
Forum Posts: 33
Member Since:
August 31, 2016
sp_UserOfflineSmall Offline
31sp_Permalink sp_Print
0

Hi Frederick,

Yes, you’re correct. The link in copper hasn’t been updated but the link that’s on the page where you purchased the board has been. 

The link is http://www.digikey.com/short/3rvd1h

Enjoy!

March 9, 2017
9:22 am
Avatar

Newbie
Forum Posts: 4
Member Since:
March 7, 2017
sp_UserOfflineSmall Offline
32sp_Permalink sp_Print
0

Thank you, Darren.  One other question, what are the 110k resistors that are by the outputs on the board for?  I do not see them on the schematic.

Thanks again.

March 9, 2017
10:40 am
Avatar
Admin
Forum Posts: 33
Member Since:
August 31, 2016
sp_UserOfflineSmall Offline

Hi Frederick,

The original schematic that Jim posted is not the exact schematic of this PCB board. From my understanding, the schematic that was posted in copper was pulled from the LM4562 datasheet that the CNC is based on. 

Since I didn’t design this preamp, I’ll have to guess what the designers intent here was since there are a couple of reasons why one might use a resistor in this location. 

The two main explanations that are jumping out to me have to deal with cutoff frequency and leakage. 

If we assume that the line preamp after the phono stage has a 100K input impedance, placing ~100K here increases the cutoff frequency to around 3Hz . This can help reduce rumble – very low frequencies that sometimes a tonearm that is “bouncing” around can produce (perhaps because of a warped record). When playing certain records, you may have seen this before when your woofers move in and out very slowly. This uses unwanted power from your amplifier and can compromise SQ.  

It also can help with leakage from the capacitor.  Capacitors are far from perfect. They have a leakage resistance that creates a voltage divider with the input resistance. Depending on this divider and the DC voltage before the cap, there can actually be DC after the cap. Lowering the input resistance decreases the DC that’s leaking. I’d expect the DC offset to be fairly low even before the cap, so leakage is not a large concern unless an op amp rails out. 

Hope this helps4_gif

March 9, 2017
2:54 pm
Avatar

Newbie
Forum Posts: 4
Member Since:
March 7, 2017
sp_UserOfflineSmall Offline
34sp_Permalink sp_Print
0

Thank you, that helps a lot.  FYI, I noticed that the minimum quantity on the .1u cap added to the BOM is 2,000.  I found a similar cap 445-4756-ND with no minimum quantity.  It is the FK series rather than the FG series, but it appears that it should work for what is needed.

Thanks again.

March 10, 2017
2:37 am
Avatar

Member
Forum Posts: 12
Member Since:
January 4, 2017
sp_UserOfflineSmall Offline
35sp_Permalink sp_Print
0

Frederick, Darren,

The 110k resistor is mainly to make sure that the op amp outputs aren’t floating when nothing’s connected. They will create a high pass filter, that’s why the value is relatively high.

It does not mess with the output impedance, as it is in parallel with the output (and two resistors in parallel will always have a resistance below the lowest value resistor).

You will see more changes from the original CNC circuit, as this is my take on this textbook phono stage. The input impedance values are (imho) more useful, giving you 17k to 150k in fifteen steps. The variable gain is done in the second gain stage, after the RIAA equalization, to avoid clipping the higher frequencies.

This circuit also gives you selectable gain of 36-46 dB in four steps, and the RIAA equalizer circuit has been improved. It has lower resistance which means lower noise, it’s not as sensitive to changes in component values (even outside of normal component tolerances), and all components are standard E24 values so you won’t have to search for components that are hard to find.

Finally, even though this board is larger than the ones in my kits, it’s smaller than the CNC boards. It’s even got ground fill, and it will slide into a B0905 enclosure without having to do any drilling.

The design files for my power supply are available here: https://www.muffsy.com/blogs/post/Make-Your-Own-Muffsy-Power-Supply/

It’s fairly easy to etch yourself, as there’s very little copper that needs removing. Alternatively, order it from the fab house of your choice.

March 13, 2017
3:32 pm
Avatar

Newbie
Forum Posts: 4
Member Since:
March 7, 2017
sp_UserOfflineSmall Offline
36sp_Permalink sp_Print
0

Thank you very much, skrodahl.  I had been thinking of using batteries, but you’ve piqued my interest with your ps.  I will give it a try on a stripboard or protoboard (this one looks interesting – https://www.tindie.com/products/SpeedyLab/prototype-pcb-board-with-ground-plane-3/

March 14, 2017
9:55 am
Avatar

Member
Forum Posts: 12
Member Since:
January 4, 2017
sp_UserOfflineSmall Offline
37sp_Permalink sp_Print
0

Cool tip about the protoboard :)

May 14, 2017
6:51 pm
Avatar

Community Leader
Forum Posts: 616
Member Since:
March 2, 2012
sp_UserOfflineSmall Offline

After months of vacations, family matters, and other get-in-the-way-of-life annoyances, I finally wired the board to my chassis and gave it a whirl on my 30 year-old turntable and cartridge resurrected from the closet.

And it worked!  Just fine.  My albums are buried in another closet, so I was stuck testing with the one 45 I had laying around (Olivia Newton-John “Physical”) It sounded fine.

I still have to make the handle for the chassis, as the batteries are heavy,  And the #1 pin on the Burr-Brown OPA2134 wasn’t in the center, so I had to guess on how to put that in, but seeing as how I had a 50-50 chance of getting it right, I guessed right the first time.

Here are pics.  Turntable is a Denon DP-52F that hadn’t been used in 30 years, coupled with a Grace F-8 cartridge of same vintage.  Worked perfectly (who’d a thunk?)

Turning the unit on and off did overload the BHK 250 and got it into rapid-blink mode from the transient spike, but all is well after a reset.  Just have to remember to turn volume down when powering the preamp on and off.

Now about those old records . . . .

DSC_0443.JPGImage Enlarger

DSC_0445.JPGImage Enlarger
DSC_0447.JPGImage Enlarger
DSC_0446.JPGImage Enlarger

Edit:  Oh, and I did change the output RCA jacks from the grounded versions to that Darren had recommended to the insulated ones as my chassis would have grounded everything together in multiple points (you don’t want that to as to prevent ground loops) so I got some insulated jacks from Mouser.  And that took a few weeks, too.

sp_PlupAttachments Attachments
August 4, 2017
6:17 pm
Avatar

Newbie
Forum Posts: 1
Member Since:
August 3, 2017
sp_UserOfflineSmall Offline

A little off-topic, but I have stumbled here and DIY Audio is my budget.

Along with my PS IV serial # 9941 waiting to be recapped, this LM194 RIAA Ultra Low Noise Preamp has been a long-delayed project of mine that I wonder if those with better hands on experience would comment on pertaining to it’s possible performance and sonic capabilities for it’s age, alongside the CNC, it’s derivatives, or the PS IV.

Many thanks in advance

Craig

LM194.1.pngImage Enlarger

LM194.2.pngImage Enlarger

sp_PlupAttachments Attachments
September 8, 2017
2:51 pm
Avatar

Community Leader
Forum Posts: 616
Member Since:
March 2, 2012
sp_UserOfflineSmall Offline
40sp_Permalink sp_Print
0

Question for skrodahl:  I want to digitize my Dad’s old CD-4 record collection so that I can play the CD4 records without any further wear on the discs.  I’d use the PSAudio NuWave Phono Converter (NPC) to digitize the CD-4 sum channel (front audio level) + the 18-45  kHz carrier (rear) into 2 digital channels at 176 kHz, then for playback send that into a DAC who’s analog output goes into the ‘phono’ input of the CD-4 decoder.

If I take the output of the cartridge and connect it directly to the analog input of the NPC, then there’d be an impedance mismatch between the cartridge and the NPC.  So I need a phono stage with input impedance matching.

Here’s the rub: When playing through a normal phono stage like the CNC before the carrier + sum is sent to the NPC, the RIAA circuit in the CNC will convert the sum channel  bass to higher levels.  I don’t want it to do that because during playback through the CD-4 decoder it will have another RIAA correction.  So when I record the record into digital I need a phono stage that has no RIAA circuit.

If I wanted to use your CNC circuit, would I take out the 47 nf, 68 nf, 130 ohm and 2.2 k ohm resistors to eliminate the RIAA?  Should the 16 k ohm resistor on the output of the first op amp be changed? If so, what would be a good value?

Thanks.

–SSW

CNC-Schematic-1.pngImage Enlarger

sp_PlupAttachments Attachments
September 9, 2017
5:18 pm
Avatar

Member
Forum Posts: 12
Member Since:
January 4, 2017
sp_UserOfflineSmall Offline

Hi Streets Still Works,

It sounds to me that you are looking for a high quality gain stage. One without the RIAA equalisation, that gives you the desired input impedance and the gain.

The Muffsy MC Head Amp is such a candidate, please contact me on the Muffsy web page for more info on the modifications needed

EDIT: The CNC without the RIAA filter will give you way to much gain. It can be used with modifications though. You need to remove everything between the two op amps, including the 16k resistor, then connect the output on the first op amp directly to the input of the second one. Change the 100 ohms resistors in the first gain stage with 360 ohms, and you’ll get the right amplification.

September 10, 2017
8:56 am
Avatar

Community Leader
Forum Posts: 616
Member Since:
March 2, 2012
sp_UserOfflineSmall Offline
42sp_Permalink sp_Print
0

Right, it’s more of a high quality gain stage with impedance matching.

Good catch on the gain on the first stage, thanks very much. I wouldn’t have thought of that. 

Also, the output of the CNC stage has that 110k ohm resistor that you don’t show here on the Muffsy circuit.  Would that be a help or hindrance to the 30 to 50 kHz rear carrier channels?  Or wouldn’t it even matter?

Thanks.

–SSW

September 10, 2017
3:12 pm
Avatar

Member
Forum Posts: 12
Member Since:
January 4, 2017
sp_UserOfflineSmall Offline
43sp_Permalink sp_Print
0

This circuit (if you disregard the RIAA filter) doesn’t cap the high frequencies in any way.

The resistor was discussed earlier in this thread, it’s just a precaution to make sure the op amps behave right if there’s nothing connected to the output. Feel free to leave it out.

September 11, 2017
9:08 am
Avatar

Community Leader
Forum Posts: 616
Member Since:
March 2, 2012
sp_UserOfflineSmall Offline
44sp_Permalink sp_Print
0

Looks like I’ll have to get another CNC board, then, and leave out the RIAA parts and short out the  16 k ohm.  I still have two spare op amps from my first CNC build.  Is Darren Myers still around?

–SSW

September 12, 2017
3:35 pm
Avatar
Admin
Forum Posts: 33
Member Since:
August 31, 2016
sp_UserOfflineSmall Offline
45sp_Permalink sp_Print
0

Hi SSW,

I’m still here! 

I had a question about your set up – whats the gain associated with the RIAA correction in the CD-4?

Is there anything else I can help you with?

If you are interested in another board, you can purchase one here: http://www.psaudio.com/product…..ono-board/

-Darren

September 12, 2017
7:42 pm
Avatar

Community Leader
Forum Posts: 616
Member Since:
March 2, 2012
sp_UserOfflineSmall Offline
46sp_Permalink sp_Print
0

Darren Myers said
I had a question about your set up – whats the gain associated with the RIAA correction in the CD-4?

Not a clue. But remember the “CD” in CD4 stands for “Compatible Discrete” which means that any 2-channel system at the time could play a CD4 record (assuming their needle was able to track the 44 kHz groove) and get the same audio played through their system as the stereo version of the record  because the 2 low frequency  channels were recorded with LF +LR & RF +RR so they’d get a full playback.  It’s only when the rear high frequency carrier signals (LF-LR, RF -RR) were added/subtracted properly would one then get individual LF, RF, LR, & RR channels.

But to answer your question I’m guessing that it’s the exact same as any standard RIAA gain because it had to play back on a ‘standard’ stereo system.

–SSW

September 13, 2017
7:29 am
Avatar

Community Leader
Forum Posts: 616
Member Since:
March 2, 2012
sp_UserOfflineSmall Offline
47sp_Permalink sp_Print
0

skrodahl said

EDIT: The CNC without the RIAA filter will give you way to much gain. It can be used with modifications though. You need to remove everything between the two op amps, including the 16k resistor, then connect the output on the first op amp directly to the input of the second one. Change the 100 ohms resistors in the first gain stage with 360 ohms, and you’ll get the right amplification.  

Skrodahl, should I get some higher resistance values to put there  just in case the gain is too high with the 360 ohm?  I don’t know a lot about analog circuits, but it looks like that resistor sets the gain of that stage by bleeding off a certain amount of the feedback signal.  So the higher the resistor, the more feedback is applied to the subtracting input of the op amp.  I was thinking of getting a 400 ohm & 500 ohm resistor to put there in case the gain is too high for the input to the NPC.  Of course, there’s the gain setting resistors on the second op amp, but seeing as how the resistors are 10 cents apiece I might as well include them in the order and have them as backup.

Thanks!

–SSW

September 13, 2017
3:15 pm
Avatar

Member
Forum Posts: 12
Member Since:
January 4, 2017
sp_UserOfflineSmall Offline
48sp_Permalink sp_Print
0

I must apologise for being too quick. The resistor value should be 1k3 (1300) ohms.

Using this value lowers the gain by the same amount as the RIAA filter would. This means that the gain switch settings will be the same as before, at 1 kHz since there’s no equalisation applied. Use a lower resistor value for more gain, or a higher resistor value for less gain.

Here’s the gain calculation for the first gain stage, before the change: http://mustcalculate.com/elect…..38;r2=3300

The attenuation of the RIAA filter (at 1 kHz) is ~20 dB, so you’d need about 20 dB less gain. 1k3 ohms gives you exactly that: http://mustcalculate.com/elect…..38;r2=3300

My first try was a brainf@rt, I aimed for 20 dB amplification instead of 20 dB lower than before.

Any resistor between 1k and two 2k ohms will get you more or less where you’d want to be, assuming that the cartridge is an MM with 3-5 mV output.

It might be the case that your software is applying the equalisation passively, which would mean that you’d want to keep the original total gain in the circuit. I’d advise against it as it most likely would make your CNC clip the higher frequencies. Let your software apply the equalisation and then amplify (or I guess it would be called normalise) the signal in the software afterwards.

September 13, 2017
4:45 pm
Avatar

Community Leader
Forum Posts: 616
Member Since:
March 2, 2012
sp_UserOfflineSmall Offline
49sp_Permalink sp_Print
0

Of course this comes an hour after I submitted the order to DigiKey for the 360 ohm parts . . .

Sigh.  OK, I’ll figure something out.  Do these resistors have to be 1% or is 5% from the local store good enough?

Oh, and even if I do sound like I’m complaining, I’m really not. Many thanks for all your help.  I couldn’t have managed without it.

–SSW

September 13, 2017
11:21 pm
Avatar

Member
Forum Posts: 12
Member Since:
January 4, 2017
sp_UserOfflineSmall Offline

Sorry man,

Always use 1% tolerance resistors, unless you need even better or if you need power resistors. Metal film is less noisy, and resistors don’t cost much anyway.

Besides, you don’t want the possibility of having 10% difference (if one is +5% and the other is -5%) in resistor values for the gain in each channel.

No permission to create posts