Hi ,
Has anyone had any experience of this kit. I am very interested in building this as it would be a big step into the Digital arena for me. As I would like to build a very decent sounding machine, I thought this might be a nice project before taking on something like the CD-PRO2.
Would really appreciate a response on this one.
Many Thanks.
Brownlow
Has anyone had any experience of this kit. I am very interested in building this as it would be a big step into the Digital arena for me. As I would like to build a very decent sounding machine, I thought this might be a nice project before taking on something like the CD-PRO2.
Would really appreciate a response on this one.
Many Thanks.
Brownlow
one is metal, one is plastic, electronic is same quality level, one is for robot loading / top load, the other for manual load/ front load/ consumer elektronics
CDM 12
Hi!
I have purchased the CDM 12, digital out (only), and remote-controlled.
I chose it over the "Pro" because of the drawer-mecanism.
As I already have a turntable, I did not want another top-loader.
Time will show what it is good for, -if (hopefull🙂) it can replace my Sonic Frontiers SFT-1 drive...
Arne K
NORWAY
And I have the shematics...e-mail me or SteinMusic for a copy.
Hi!
I have purchased the CDM 12, digital out (only), and remote-controlled.
I chose it over the "Pro" because of the drawer-mecanism.
As I already have a turntable, I did not want another top-loader.
Time will show what it is good for, -if (hopefull🙂) it can replace my Sonic Frontiers SFT-1 drive...
Arne K
NORWAY
And I have the shematics...e-mail me or SteinMusic for a copy.
Hi Brownlow,
Works like a champ but I still prefer vinyl over polycarbonate 😀
I use the DF1704/PCM1704 combo. The low jitter clock (thanks Guido!) is positioned next to the DACs and is feed back to the drive. The CDM12 drive electronics are on an large old style pcb which makes modification to the circuit simple.
I do the I/U conversion by use of a simple 22R resistor. The output stage consists of a step up transformer followed by a single tube stage and an 1:1 output transformer (picture shows an older version with different output stage). My $2000 commercial CD player has not spun a single CD since I first listened to this mess of parts.
Manfred
I have built a prototype CD player using this drive:Brownlow said:Has anyone had any experience of this kit. I am very interested in building this as it would be a big step into the Digital arena for me.

An externally hosted image should be here but it was not working when we last tested it.
Works like a champ but I still prefer vinyl over polycarbonate 😀
I use the DF1704/PCM1704 combo. The low jitter clock (thanks Guido!) is positioned next to the DACs and is feed back to the drive. The CDM12 drive electronics are on an large old style pcb which makes modification to the circuit simple.
I do the I/U conversion by use of a simple 22R resistor. The output stage consists of a step up transformer followed by a single tube stage and an 1:1 output transformer (picture shows an older version with different output stage). My $2000 commercial CD player has not spun a single CD since I first listened to this mess of parts.

Whoa!
Manfred,
That is some serious stuff you've got there...What DACs are you using (or maybe the real question is, what DACs are good)?
We know good chips cost money but this would give me an idea as to what I will need.
I thought it was time that I had a decent CD player. From what I've read so far this looks to be like some good kit.
Thanks for all contributions as usual, what would I do without this forum?!
Cheers.
Brownlow
Manfred,
That is some serious stuff you've got there...What DACs are you using (or maybe the real question is, what DACs are good)?
We know good chips cost money but this would give me an idea as to what I will need.
I thought it was time that I had a decent CD player. From what I've read so far this looks to be like some good kit.
Thanks for all contributions as usual, what would I do without this forum?!
Cheers.
Brownlow
Re: Whoa!
I think he already stated that he's using BB PCM1704 DACs, which is what I'm using too. Manfred, were you not tempted to go non oversampling route?😉
I would be more interested in a question, what output transformers are you using? Bernhard praised it so much, that your setup is almost legendary in here.😉
Brownlow said:Manfred,
What DACs are you using (or maybe the real question is, what DACs are good)?
I think he already stated that he's using BB PCM1704 DACs, which is what I'm using too. Manfred, were you not tempted to go non oversampling route?😉
I would be more interested in a question, what output transformers are you using? Bernhard praised it so much, that your setup is almost legendary in here.😉
Re: Whoa!
Brownlow,
Manfred
Brownlow,
As I already have said I use the Burr-Brown (now TI) DF1704 digital filter and a pair of BB PCM1704 DACs. You may want to got to Google and type in PCM1704. You will have no Problem finding the data sheet and many comments on this chip.Brownlow said:What DACs are you using (or maybe the real question is, what DACs are good)?

Re: Re: Whoa!
Hi Peter,
helps with this kind of heretical thoughts! I guess I'm too much of an engineer to throw away all I have learnt about digital signal processing. Don't get me wrong folks, I'm not saying that non oversampling DACs cannot sound good but it's simply not my cup of tea.
I use a shunt regulator to generate the plate voltage (only 70V) for the DL91s. The filaments of the DL91s are heated by current sources and the bias voltage for the tube is coming from a lithium cell.
There is no analog reconstruction filter. The 8 times oversampling filter shifts digital artefacts up to about 350kHz and the signal transformers do a good job attenuating the junk up there.
Manfred
Hi Peter,
From time to time I feel tempted to try non oversampling by reading glowing reports on the net. Self-flagellationManfred, were you not tempted to go non oversampling route?

The analog part looks like this: 22R resistor for I/U conversion, Sowter 8347 1:18.7 step up transformer, DL91 directly heated pentode connected as pseudo triode driving a Lundahl LL1635 transformer (www.lundahl.se). The LL1635 originally is an interstage transformer but I use it at the output. The tube has a plate resistance of about 3K to which the LL1635 adds about 2K of copper resistance giving a total output impedance of around 5K. Some may consider 5K to be too high but it works fine for me. This may be different if one has to drive long high capacitance interconnects.I would be more interested in a question, what output transformers are you using?
I use a shunt regulator to generate the plate voltage (only 70V) for the DL91s. The filaments of the DL91s are heated by current sources and the bias voltage for the tube is coming from a lithium cell.
There is no analog reconstruction filter. The 8 times oversampling filter shifts digital artefacts up to about 350kHz and the signal transformers do a good job attenuating the junk up there.
He's too kind 🙂 but I also have to give credit to my friends Thomas Mayer and Guido Tent. They provided some of the essential ingredents (i.e.concepts). All I had to do was cooking the meal.Bernhard praised it so much, that your setup is almost legendary in here.


Hallo Manfred!
welcome back
Uhhm, semantic correction, i am not kind, i am the same kind of monster. 🙂
Guido Tent also is around here now and then. Guido? You are out there?
I know Thomas Mayer's implemenation and must say, i am impressed, he is crazier than me, his digital setup comes close to the insane TT design of my diploma thesis; considering it is merely digital, it vastly outperforms it, and it exists and sounds simply terrific.
But your setup is more to my personal sonic taste and it has a fine balance of effort vs sonics. A gorgeous balance.
Thats why i am raving, anyone can do it for $5000, a master who does it for $50.
I chose numbers arbitrary and i am exaggerating to illustrate my point. 😀
I am convinced, planet10 will let you use his frugalphile(tm) label with out license fees 🙂, just observing your choice of CD drive and tube
welcome back

before i found that line, i thot, heck, can a digital camera have such heavy THD? 🙂... an 1:1 output transformer (picture shows an older version with different output stage).
Uhhm, semantic correction, i am not kind, i am the same kind of monster. 🙂
Guido Tent also is around here now and then. Guido? You are out there?
I know Thomas Mayer's implemenation and must say, i am impressed, he is crazier than me, his digital setup comes close to the insane TT design of my diploma thesis; considering it is merely digital, it vastly outperforms it, and it exists and sounds simply terrific.
But your setup is more to my personal sonic taste and it has a fine balance of effort vs sonics. A gorgeous balance.
Thats why i am raving, anyone can do it for $5000, a master who does it for $50.
I chose numbers arbitrary and i am exaggerating to illustrate my point. 😀
I am convinced, planet10 will let you use his frugalphile(tm) label with out license fees 🙂, just observing your choice of CD drive and tube

Re: Re: Steinmusic cdm12.1
Hi Manfred,
What range of resistors did you try for I/V conversion and what was the affect?
Thanks,
Ian
mhuber said:Hi Brownlow,
I have built a prototype CD player using this drive:![]()
An externally hosted image should be here but it was not working when we last tested it.
Works like a champ but I still prefer vinyl over polycarbonate 😀
I use the DF1704/PCM1704 combo. The low jitter clock (thanks Guido!) is positioned next to the DACs and is feed back to the drive. The CDM12 drive electronics are on an large old style pcb which makes modification to the circuit simple.
I do the I/U conversion by use of a simple 22R resistor. The output stage consists of a step up transformer followed by a single tube stage and an 1:1 output transformer (picture shows an older version with different output stage). My $2000 commercial CD player has not spun a single CD since I first listened to this mess of parts.
Manfred
Hi Manfred,
What range of resistors did you try for I/V conversion and what was the affect?
Thanks,
Ian
Hi Ian,
with my own unit I have only tried a value of 22R. This was because 22R gives a good match for the Sowter 8347 transformer.
I have built an other unit for a friend which uses a 47R resistor, a third order passive L-C-L Bessel low pass and a single tube stage for amplification. This one sounds quite different but It is a completely different setup, not just a changed resistor value.
In general there are two things to take into consideration:
The D/A converter's output is protected by a pair of diodes against electrostatic discharge etc. These diodes will start to conduct when the output reaches about 600mV. We have to keep the I/U converting resistor low enough to prevent this from happening. In my case the D/A converter outputs max. +-1.25mA of current and the resistor is 22R. This means that the peek output voltage will be 22R*1.25mA=27.5mV. I guess this is a save bet. 😀
The other thing to consider is that I/U conversion done by a simple resistor will generate some 2nd order distortion products. This is because the output of the converter is not a real current source.
The output resistance of the converter chip at max output level is Vref/(2*Iout_max). In the case of the PCM1704 this is 5V/2.5mA=2000R. At min output level all switches in the R2R ladder network are open and there is no current flowing at all. The output resistance is indefinitely high. Let's see what this means for the outout current:
The min output current is Imin=-1.25mA. This is the current set by the
internal bias current source. There is no current through the R2R network in this case.
The max. output current is Vref divided by the sum of the resistances of the R2R network and the I/U conversion resistor less the bias current of 1.25mA:
Imax=Vref/(2000R+22R)-1.25mA=1,2228mA
This is 2,2% of deviation from the ideal of 1.25mA. The 2nd order distortion calculates as:
K2(%) = 100 * (Imax+Imin)/2/(Imax+Imin)
In our case:K2 = 100*(1.2228-1.25)/ 2/(1.2228+1.25) = 0,55%
This is quite a bit of distortion but it is purely 2nd. order and there is a near linear relationship of distortion and level. The calculation was done under the assumption of max output level. If your CD is recorded at an average level of -20dB then there will also be about 20dB less distortion. In our case this means 0,055% of K2. Good enough for me.
I hope this helps.
I also hope I didn't make too many errors
Regards
Manfred
what range of resistors did you try for I/V conversion and what was the affect?
with my own unit I have only tried a value of 22R. This was because 22R gives a good match for the Sowter 8347 transformer.
I have built an other unit for a friend which uses a 47R resistor, a third order passive L-C-L Bessel low pass and a single tube stage for amplification. This one sounds quite different but It is a completely different setup, not just a changed resistor value.
In general there are two things to take into consideration:
The D/A converter's output is protected by a pair of diodes against electrostatic discharge etc. These diodes will start to conduct when the output reaches about 600mV. We have to keep the I/U converting resistor low enough to prevent this from happening. In my case the D/A converter outputs max. +-1.25mA of current and the resistor is 22R. This means that the peek output voltage will be 22R*1.25mA=27.5mV. I guess this is a save bet. 😀
The other thing to consider is that I/U conversion done by a simple resistor will generate some 2nd order distortion products. This is because the output of the converter is not a real current source.
The output resistance of the converter chip at max output level is Vref/(2*Iout_max). In the case of the PCM1704 this is 5V/2.5mA=2000R. At min output level all switches in the R2R ladder network are open and there is no current flowing at all. The output resistance is indefinitely high. Let's see what this means for the outout current:
The min output current is Imin=-1.25mA. This is the current set by the
internal bias current source. There is no current through the R2R network in this case.
The max. output current is Vref divided by the sum of the resistances of the R2R network and the I/U conversion resistor less the bias current of 1.25mA:
Imax=Vref/(2000R+22R)-1.25mA=1,2228mA
This is 2,2% of deviation from the ideal of 1.25mA. The 2nd order distortion calculates as:
K2(%) = 100 * (Imax+Imin)/2/(Imax+Imin)
In our case:K2 = 100*(1.2228-1.25)/ 2/(1.2228+1.25) = 0,55%
This is quite a bit of distortion but it is purely 2nd. order and there is a near linear relationship of distortion and level. The calculation was done under the assumption of max output level. If your CD is recorded at an average level of -20dB then there will also be about 20dB less distortion. In our case this means 0,055% of K2. Good enough for me.
I hope this helps.
I also hope I didn't make too many errors

Regards
Manfred
Hi Manfred,
Thanks for the detailed response, do you have documentation for the CDM12?
Regards,
Ian
Thanks for the detailed response, do you have documentation for the CDM12?
Regards,
Ian
Hi Ian,
I guess Holger Stein will send you the schematic if you ask nicely.
Regards
Manfred
It came with some information on hook-up etc. but that wasn't much.iansmith14 said:
Thanks for the detailed response, do you have documentation for the CDM12?
I guess Holger Stein will send you the schematic if you ask nicely.
Regards
Manfred
Clock Frequency
Hi Manfred,
May I ask what is the masterclock frequency the CDM12 uses?😕
Last Friday I saw a expensive Euro 4000 costing transport using the CDM12 with the crystal hung on the microprocessor.😕
Hi Manfred,
May I ask what is the masterclock frequency the CDM12 uses?😕
Last Friday I saw a expensive Euro 4000 costing transport using the CDM12 with the crystal hung on the microprocessor.😕
Re: Clock Frequency
Manfred
11.2896MHzElso Kwak said:May I ask what is the masterclock frequency the CDM12 uses?
You know, the high end industry considers a 4000 Euro transport being a "mid price" product. There has to be some room for improvements 😀Last Friday I saw a expensive Euro 4000 costing transport using the CDM12 with the crystal hung on the microprocessor.😕

- Status
- Not open for further replies.
- Home
- Source & Line
- Digital Source
- Steinmusic cdm12.1