Hi Guys,
Finally got my hands on a CD2 thats mine and I want to change the output stage. I want to remove all the opamps and join all the connections and put it in a seperate preamp like class A.
The CD2 has 4x TDA1541 and I believe each DAC has 4ma output. Total = 16ma. I'm not sure what magnification the datasheet is, it has a 1.8k resistor across it.
I don't want valves as its beyond me and high voltages, transformers would be nice but its alien technology to me. So I was considering preamp boards, there seems to be plently from China that are Naim or other based that offer some form of magnification. Has anyone tried any of them with TDA1541 or reccomend anything? It is not enough magnification?
I plan to build an external housing with its own PSU. Thanks in advance for any info offered
Finally got my hands on a CD2 thats mine and I want to change the output stage. I want to remove all the opamps and join all the connections and put it in a seperate preamp like class A.
The CD2 has 4x TDA1541 and I believe each DAC has 4ma output. Total = 16ma. I'm not sure what magnification the datasheet is, it has a 1.8k resistor across it.
I don't want valves as its beyond me and high voltages, transformers would be nice but its alien technology to me. So I was considering preamp boards, there seems to be plently from China that are Naim or other based that offer some form of magnification. Has anyone tried any of them with TDA1541 or reccomend anything? It is not enough magnification?
I plan to build an external housing with its own PSU. Thanks in advance for any info offered
Attachments
For example this board offers 11x magnification
Core line:BRYSTON DOA33 module, full DC design, no input and output capacitors, the sound is super transparent.
Quiescent current: voltage class 10mA, output class 20mA class A current, the core of good sound.
Output noise: 0.2mV, output midpoint: 0.4mV (hard indicator is super good).
Magnification: 11 times;
Supply voltage: DC ± 33V;
Transformer power: 30w;
Printing plate size: 103*131mm; (printing plate is not small)
Pre-potentiometer: 100kA;
Core line:BRYSTON DOA33 module, full DC design, no input and output capacitors, the sound is super transparent.
Quiescent current: voltage class 10mA, output class 20mA class A current, the core of good sound.
Output noise: 0.2mV, output midpoint: 0.4mV (hard indicator is super good).
Magnification: 11 times;
Supply voltage: DC ± 33V;
Transformer power: 30w;
Printing plate size: 103*131mm; (printing plate is not small)
Pre-potentiometer: 100kA;
The TDA1541 is a current-output DAC. It's outputs have only 25mV of compliance, so use the standard IV converter circuit from the datasheet or they will not perform at all.
It's too complicated for me to change the output. Unfortunately I will have to stick with original opamps. I know opa627 is better but £30 x 8 opamps= £240 just too much
I've not heard 627's (they were too expensive for me too!). I've put OPA134's in mine, a better clock and a few PSU mods (diodes and cap upgrades). If you can put some heatsinks on the DAC chips and improve the cooling air throughput, that’s a good idea also. It's a very underated player (if a bit ugly) and a real giant killer.
PS.
I've put connectors on my DAC main board (the one that the 4 DAC boards fit onto) for ease of modding, they work well.
PS.
I've put connectors on my DAC main board (the one that the 4 DAC boards fit onto) for ease of modding, they work well.
Attachments
Last edited:
Very nice! I was looking into OPA1655 but I know already nothing beats OPA627, I've used them before. I was drooling over the possibility of wiring all the output together and using a discrete opamp at the end of it!
I created this and double checked it 5 times so I hope its correct but oddly look at the output of the DAC it goes straight to the output via 2.2k resistors parallel to the opamp output. This could be normal but its strange to me lol
I created this and double checked it 5 times so I hope its correct but oddly look at the output of the DAC it goes straight to the output via 2.2k resistors parallel to the opamp output. This could be normal but its strange to me lol
Attachments
Don't mod it. It is not sure your shematic is rigth. Seems to me there is I/V + reconstruction filter (2k2 serie with 6n8 in between to ground?). Or maybe it is just one 5534 per TDA1541A. Without shematic don't play with it !
The Signetic 5534 are the best 5534 op amps made and keepers....ultra low noise. You could at least change the I/V for something that is drop replacement for the 5534 but much beter for the I/V task, but we don't know the shematic really, so...
op1655, op828 (better than the 627 for much less), or easier : LM6171, but on a socket it will oscillate. OP1655 is what I'd choose, it has some nice feature forthat I/V task and work fine here. Of course some needs soic8 to dip8 8 adaptators. You have to check the voltage compatibility as well...
The Signetic 5534 are the best 5534 op amps made and keepers....ultra low noise. You could at least change the I/V for something that is drop replacement for the 5534 but much beter for the I/V task, but we don't know the shematic really, so...
op1655, op828 (better than the 627 for much less), or easier : LM6171, but on a socket it will oscillate. OP1655 is what I'd choose, it has some nice feature forthat I/V task and work fine here. Of course some needs soic8 to dip8 8 adaptators. You have to check the voltage compatibility as well...
Yes look at my drawings it's 4x ne5534 per channel. 8 in total. They are also Signetic.
I have used opa627 and opa1655 before. Never Lm6171. I will look into that.
The problem is that if I use a socket for opa1655 it will hit the heatsink on the TDA1541.
Is passive transformer any good? I'm not a fan of valves
I have used opa627 and opa1655 before. Never Lm6171. I will look into that.
The problem is that if I use a socket for opa1655 it will hit the heatsink on the TDA1541.
Is passive transformer any good? I'm not a fan of valves
Not really or very expensive, hand made on spec, will anyway need some active for dynamic. The usual culpritt like Sowter is a so-so choice, typical diy trap imho !
The shematic seems again a little odd to me, a mistake when you draw it I surmise. I bet more for an I/V stage plus some reconstruction filter after.
LM6171 : no socket, direct soldering on pcb. DIP8
Op1655 : no socket, just the adapter pcb direct soldered as "Brown Dog"s adapter for illustration, 1.8 mm thick 🙂
For the I/V only and you keep the Signetics for the filter after (if I am rigth).
In your shoes the 1655 is worthing the extra effort imo.
heaatsink is less important than the op amp choice for the sound. Just don't use your DAC 12 hours. I never lost a TDA1541A in 30 years because of heat.
The shematic seems again a little odd to me, a mistake when you draw it I surmise. I bet more for an I/V stage plus some reconstruction filter after.
LM6171 : no socket, direct soldering on pcb. DIP8
Op1655 : no socket, just the adapter pcb direct soldered as "Brown Dog"s adapter for illustration, 1.8 mm thick 🙂
For the I/V only and you keep the Signetics for the filter after (if I am rigth).
In your shoes the 1655 is worthing the extra effort imo.
heaatsink is less important than the op amp choice for the sound. Just don't use your DAC 12 hours. I never lost a TDA1541A in 30 years because of heat.
Last edited:
Yes the brown dog seems small enough. The one I was looking at the legs are too extended.
Interesting I found this online:
i found the lm6171 to be superior in nearly every respect: detail, pace, timbre, accuracy, neutrality.
How would you rate that opa1655? I used this before and I thinks it's very good too but opa627 is also very good. Discrete beats both by miles!!
Tell me more about passive transformer output. I was looking at Sowter too because they are in UK. You say passive is not good?? Would it be better to put the transformer after the opamps?? It's nearly £250! Extra cost!!
I studied the drawing for output several times already I'm 99.9% sure it accurate. It's without the de-emphasis.
Thank for the input
Interesting I found this online:
i found the lm6171 to be superior in nearly every respect: detail, pace, timbre, accuracy, neutrality.
How would you rate that opa1655? I used this before and I thinks it's very good too but opa627 is also very good. Discrete beats both by miles!!
Tell me more about passive transformer output. I was looking at Sowter too because they are in UK. You say passive is not good?? Would it be better to put the transformer after the opamps?? It's nearly £250! Extra cost!!
I studied the drawing for output several times already I'm 99.9% sure it accurate. It's without the de-emphasis.
Thank for the input
prove is in the pudding. 🙂
Seems here they put most the filtering in the oversampling before it reaches the DAC to allow a smoother one after....
Seems here they put most the filtering in the oversampling before it reaches the DAC to allow a smoother one after....
That picture could be for CD3 although the board looks like CD2. I have never seen a CD2 with 5uf output only CD3. The drawing is identical and better than mine lol. Except my first cap is 6n8 and 4.7uf
The CD2, CD3 and DAC3 D/A all operate using the same principles. It is what the 16x16 refers to.
Yes absolutely. Please see my other thread in regards to this:
This chip is/was used in Cambridge CD2, CD3 and the DAC3 (I think) from my understanding the digital data signal goes into that then goes to the DAC chips.
My question is I repaired one that failed, it was the only real problem that needed addressing to get sound out. Why was it so simple? well it seems to be a very very common problem and my question today if anyone has any answers ias to why does it fail so often? Its a bit late to complain to the manufacturer since its 30 + years old but just very curious as to why those particular chips fail even before the caps!?
There seems to be...
My question is I repaired one that failed, it was the only real problem that needed addressing to get sound out. Why was it so simple? well it seems to be a very very common problem and my question today if anyone has any answers ias to why does it fail so often? Its a bit late to complain to the manufacturer since its 30 + years old but just very curious as to why those particular chips fail even before the caps!?
There seems to be...
- dj_holmes
- Replies: 29
- Forum: Digital Line Level
There is a guy on here that makes and sells very good transformers for DAC outputs.Is passive transformer any good? I'm not a fan of valves
Hello everyone who looking for Sound!

After years of expirience the output transformers for current and voltage output DACs have been developed. Transformers tested with following DACs: ES90x8S/K2M/PRO/Q2M, AK4x9x, PCM56/58/63, PCM1702/04, AD1862/65, PCM1792, PCM1794A, TDA1541/43, DSC dac and others. Speaking in general, transformers can be suited for any current or voltage DAC both in SE or PP modes.
These transformers are made on high permeable cores, have the low DC resistance at desirable inductive impedance to achieve the best frequency response of the DAC. Transformers...

After years of expirience the output transformers for current and voltage output DACs have been developed. Transformers tested with following DACs: ES90x8S/K2M/PRO/Q2M, AK4x9x, PCM56/58/63, PCM1702/04, AD1862/65, PCM1792, PCM1794A, TDA1541/43, DSC dac and others. Speaking in general, transformers can be suited for any current or voltage DAC both in SE or PP modes.
These transformers are made on high permeable cores, have the low DC resistance at desirable inductive impedance to achieve the best frequency response of the DAC. Transformers...
- bisesik
- Replies: 982
- Forum: Vendor's Bazaar
- Home
- Source & Line
- Digital Source
- Cambridge CD2 new output stage