Tried this and it resulted in return of "LSB noise" and whine. Even more so than the orig 22k + 33k before cleaning up soldering joints, etc . I haven't yet returned back to orig 22k + 33k, where I had it singing almost perfectly . So maybe a soldering compromise when modding in 2x11k. I suspect that 1545 is capable of running just fine with the DEFAULT datasheet suggestion. But it's tricky to tweak given the DF1704 is sending such "high throughput" data (?????). So traces and power connections, ground and loops have to be just right.
Not sure why the DS suggests the 2 * 11k mode? Maybe for certain I/V or opamp pairings. And also very possible with the Vref (7) attached to the OPA non-inverting input. But not the Rudolf I/V I'm using.
What's your output voltage on the 1545? Turns out that the output compliance spec in the DS is wrong when the chip's programmed to deliver 2mA. In practice, at 2mA the minimum output voltage is closer to 3V than the 2V cited in the DS. The symptoms of running beyond the DAC's compliance sound like heavy noise modulation. Not sure I've ever had 'whine' though.
Problem solved, with the 2x11k still in place.
In the photo, you'll note the addition of two new high-quality Pana caps (1 uF) across the TDA1545 Vcc / gnd pins. These are new to the design.
Before, with the 22k/33k, adding any caps across these pins resulted in noise. But I had a hunch -- a "Dream Within a Dream" you might say -- and like magic, noiseless, whine-less sonics again.
"By what eternal streams ..."
In the photo, you'll note the addition of two new high-quality Pana caps (1 uF) across the TDA1545 Vcc / gnd pins. These are new to the design.
Before, with the 22k/33k, adding any caps across these pins resulted in noise. But I had a hunch -- a "Dream Within a Dream" you might say -- and like magic, noiseless, whine-less sonics again.
"By what eternal streams ..."
As a conclusion to the DF1704/TDA1545 experiment, the final verdict is that 3-line serial output format spat out by the 1704 is 100% compatible with the TDA1545a .
If anyone requires further proof, do the experiment yourself -- or, if I get enough requests, I can "show my work" by recording the output of the CD player (see photos above) into a Sony TRV99 analog hi8 camcorder, and upload the video. The TRV99's analog AFM (hifi stereo) pre-amp stage is superior to that found in most modern digital camcorders -- it can revel quite a bit even over streamed Internet video @ the usual 128k.
Some of the problems I noted earlier -- noise, whine, etc - -were my own mistakes incl. sloppiness in hurried veroboard build, not following DS decoupling notes, etc.
The confusion about incompatibility -- noted by others -- may be partially based on errors or generalizations in the two datasheets.
The fidelity of the DF1704/TDA1545, in that 1989 Philips CDP, with the Rudolf I/V, is excellent ... with good channel separation and imaging -- a bit of surprise given the un-used channel setup. Much more potential here for further sonic upgrades or alternative inputs, such as a dedicated clock, re-clock, cleaner power, and USB-I2S Amanero input .
If anyone requires further proof, do the experiment yourself -- or, if I get enough requests, I can "show my work" by recording the output of the CD player (see photos above) into a Sony TRV99 analog hi8 camcorder, and upload the video. The TRV99's analog AFM (hifi stereo) pre-amp stage is superior to that found in most modern digital camcorders -- it can revel quite a bit even over streamed Internet video @ the usual 128k.
Some of the problems I noted earlier -- noise, whine, etc - -were my own mistakes incl. sloppiness in hurried veroboard build, not following DS decoupling notes, etc.
The confusion about incompatibility -- noted by others -- may be partially based on errors or generalizations in the two datasheets.
The fidelity of the DF1704/TDA1545, in that 1989 Philips CDP, with the Rudolf I/V, is excellent ... with good channel separation and imaging -- a bit of surprise given the un-used channel setup. Much more potential here for further sonic upgrades or alternative inputs, such as a dedicated clock, re-clock, cleaner power, and USB-I2S Amanero input .
But it needs two TDA1545As, is that correct? In which case I'd say its 50% compatible.
The 1541 can time multiplex with simple pin tweak (table 1). I think I pretty much had the same dream Philips 1541 planners did in mid 80s. Recall earlier I did inquire about time MUX before the 1545, say via 74hc logic. But got no reply. But that was not part of the orig dream anyway, so no worries 😉
Music: Bach/Scarlatti/Mozart/Schubert: Piano Recital
Music: Bach/Scarlatti/Mozart/Schubert: Piano Recital