@wynpalmer2 - FYI I think the mouser baskets back on post 48 are corrupted. I started with the simplest bit of assembly - the SMPS filter, but the electrolytics and LED resistor are missing from that list....
https://www.diyaudio.com/forums/ana...filter-line-headphone-amps-5.html#post6409267
https://www.diyaudio.com/forums/ana...filter-line-headphone-amps-5.html#post6409267
No worries @wynpalmer2 - I should have most bits, and if something else is missing then it can always be obtained. I didn't do an exhaustive check when I ordered, I just checked a bunch of items from each list to make sure I didn't miss a whole list.
I think if you tell a mod they can change it for you
I think if you tell a mod they can change it for you
I'm a subscriber to AK and that status provides the ability to edit past posts, and I wish it was possible here. Having to invoke a mod to make a change is just poor.
The current global components shortage is affecting new builds of the preamp. I've just ordered parts for the first v3.0 build- and of course the new section is problematic as the output relays are currently unobtainium until August, so I had to find a (very expensive) replacement. There were also close to a dozen, of a total of c.60 different parts, on backorder.
The current global components shortage is affecting new builds of the preamp. I've just ordered parts for the first v3.0 build- and of course the new section is problematic as the output relays are currently unobtainium until August, so I had to find a (very expensive) replacement. There were also close to a dozen, of a total of c.60 different parts, on backorder.
Yes, even though my orders are infrequent and miniscule, even I have noticed how many parts seem to be out of stock with long lead times. I'm keeping a note of what I've missed as I go along. I know I didn't order U9, and I'm short 10uf caps as well.
We'll see how it goes, hopefully its not too bad.
Fran
We'll see how it goes, hopefully its not too bad.
Fran
I am testing out a new piece of software to go with the RME box so that i can make some measurements that I cannot otherwise make- Virtual Instrument (Virtins).
Below is a THD/GedLee metric measurement of the post phono stages in the preamp (warp filter. buffer, passive volume control, output buffer/amplifier, plus the RME DAC and ADC.
The output signal is 1v rms, and both the THD metric and the GedLee metric are identical, which means that the measures that I took to ensure that the levels of high order non linearity were acceptable was successful.
The same result applies to the phono stage itself, except the noise is much higher and the result is less impressive.
Below is a THD/GedLee metric measurement of the post phono stages in the preamp (warp filter. buffer, passive volume control, output buffer/amplifier, plus the RME DAC and ADC.
The output signal is 1v rms, and both the THD metric and the GedLee metric are identical, which means that the measures that I took to ensure that the levels of high order non linearity were acceptable was successful.
The same result applies to the phono stage itself, except the noise is much higher and the result is less impressive.
Attachments
There will be a small revision to the next board order, making it v3.1.
The updated schematic is below.
The build guide has also been updated for v3.x
Dropbox - LOMC-v3-Addendum.pdf - Simplify your life
The updated schematic is below.
The build guide has also been updated for v3.x
Dropbox - LOMC-v3-Addendum.pdf - Simplify your life
I've built and been living with 2 of Wyns boards over the last few months - an MC and a MM build. I've done quite a bit of tricking about with the power supplies, but the phono builds apart from that are as per the datasheet. I've learned quite a few things along the way, but I'd like to summarise my findings:
1. This is a high component build, but it offers things that are hard to find on other diy builds. In particular, the warp filter gives a major advantage. It really works well, and I can't think of many other offerings with it.
2. I found the power supply to matter quite a lot - although it has to be noted that I strayed from the SMPS recommended by Wyn and used a cheapo one from Aliexpress. Moving from there to a better linear supply brought the sound quality to a higher level (all in my own opinion of course). I am using a supply with supercaps now for one board, and I have tried a few linear supplies with the MM version, in fact I just fitted a VRDN one this evening and am enjoying it as I type here.
3. Design is rock solid - no DC offset issues or hot components etc My impression is that this is the kind of build that will still be playing in 10 or 20 years time.
4. Sound quality is excellent - really excellent. There's a depth to the soundstage and the ability to pick and follow individual instruments through a piece.
I don't have much of a choice of MM cartridges, in fact I'm using this board with a HOMC at present, but I am curious as to how this would sound with a decent MM cartridge. I have a NOS AT450 here that I think migh be worth a spin with it.
In short, this is a vote for this build. Yes, it has a higher parts count than many, but in turn you get things like the warp filter, the gain settings, mono switch, balanced outputs, good interface to headphone amp etc as well.
1. This is a high component build, but it offers things that are hard to find on other diy builds. In particular, the warp filter gives a major advantage. It really works well, and I can't think of many other offerings with it.
2. I found the power supply to matter quite a lot - although it has to be noted that I strayed from the SMPS recommended by Wyn and used a cheapo one from Aliexpress. Moving from there to a better linear supply brought the sound quality to a higher level (all in my own opinion of course). I am using a supply with supercaps now for one board, and I have tried a few linear supplies with the MM version, in fact I just fitted a VRDN one this evening and am enjoying it as I type here.
3. Design is rock solid - no DC offset issues or hot components etc My impression is that this is the kind of build that will still be playing in 10 or 20 years time.
4. Sound quality is excellent - really excellent. There's a depth to the soundstage and the ability to pick and follow individual instruments through a piece.
I don't have much of a choice of MM cartridges, in fact I'm using this board with a HOMC at present, but I am curious as to how this would sound with a decent MM cartridge. I have a NOS AT450 here that I think migh be worth a spin with it.
In short, this is a vote for this build. Yes, it has a higher parts count than many, but in turn you get things like the warp filter, the gain settings, mono switch, balanced outputs, good interface to headphone amp etc as well.
Hello,
Nice work, i own an hypnotized pre and it's nice, i didn't follow the latest development on audio karma though. When i red the thread i was interested to build one, a shame you remove the balanced out for function i do not need. If by any chance you still have a v2 board somewhere or plan to add again the balanced out in the next revision...
Nice work, i own an hypnotized pre and it's nice, i didn't follow the latest development on audio karma though. When i red the thread i was interested to build one, a shame you remove the balanced out for function i do not need. If by any chance you still have a v2 board somewhere or plan to add again the balanced out in the next revision...
The balanced/differential out was just a +1/-1 gain stages driving what was in effect the record and line outputs. The blocks are still there as +1 and +2 and can be wired as +1 and -1 quite easily. It doesn't look as pretty, but it works. The relay outputs etc. still work. The balance/mono controls must be removed/bypassed.
The conversion ceased to be just a pair of jumpers and it was not used by anyone that I was aware of, which is why it was removed.
I do not have any 2.0 boards left- indeed any boards at all bar a couple of headphone units- but some v3.1 boards should be arriving shortly.
The conversion ceased to be just a pair of jumpers and it was not used by anyone that I was aware of, which is why it was removed.
I do not have any 2.0 boards left- indeed any boards at all bar a couple of headphone units- but some v3.1 boards should be arriving shortly.
Yes. providing the current capacity, output impedance and noise are adequate. The design is supply type agnostic.
Noise less than about 1mv RMS over the audio band, >200mA DC current and less than 2 ohms output impedance over the audio band should do the job.
If the headphone amp is added then the output impedance should be <100mohms, and the DC current should be 1A or higher.
Noise less than about 1mv RMS over the audio band, >200mA DC current and less than 2 ohms output impedance over the audio band should do the job.
If the headphone amp is added then the output impedance should be <100mohms, and the DC current should be 1A or higher.
Wyn,
I recently received your v3.1 boards and I'm about to start assembly.
I realized that I have a few AD745-JK 8-pin DIP (tube says they were made in 1993, so definitely NOS).
My question is what is your opinion on how, in parallel, these would stack up in the MM section vs the OPA1641?
From the data sheets, the AD745 has less noise (2.9 vs 5.1), but more current noise (6.9 vs 0.8).
I recently received your v3.1 boards and I'm about to start assembly.
I realized that I have a few AD745-JK 8-pin DIP (tube says they were made in 1993, so definitely NOS).
My question is what is your opinion on how, in parallel, these would stack up in the MM section vs the OPA1641?
From the data sheets, the AD745 has less noise (2.9 vs 5.1), but more current noise (6.9 vs 0.8).
Last edited:
With the 745 noise is a more complex issue. The input current noise is still two orders of magnitude less than bipolars and at c. 7fA it wont be comparable to the voltage noise until the source impedance is c. 300kohms. However, the noise performance is degraded substantially if the impedances seen by the two input terminals differ, which they certainly will in this case. The data sheet indicates that some correlated contributor to the input stage voltage noise appears at the input terminals due to capacitive coupling. See fig. 20 in the data sheet. I would have to run simulations to determine the effect of this, but I'm not presently in a position to do so.
Another issue is that the 745 is not unity gain- but gain of +5 - stable, so the input feedback stages have to be modified a bit. This is because the input includes the 75uA TC and so it falls to a gain of 5 at c. 8kHz and progressively lower above that frequency.
Without a change the input stage will likely oscillate at several MHz.
This particular problem can be solved by increasing the minimum closed loop noise gain for the input stage to +5. This can be done by increasing the 75 ohm output Rs to c.4x the value of the resistance between the negative input and ground- or about 430 ohms.
Another issue is that the 745 is not unity gain- but gain of +5 - stable, so the input feedback stages have to be modified a bit. This is because the input includes the 75uA TC and so it falls to a gain of 5 at c. 8kHz and progressively lower above that frequency.
Without a change the input stage will likely oscillate at several MHz.
This particular problem can be solved by increasing the minimum closed loop noise gain for the input stage to +5. This can be done by increasing the 75 ohm output Rs to c.4x the value of the resistance between the negative input and ground- or about 430 ohms.
Last edited: