ezkcdude said:
Is there a simple but high quality alternative for voltage output DACs?
If the DC offset - as mentioned above - is quite easy to estimate (1/2 VCC) it should be possible to get rid of it. So something like a mosfet stage should be a viable way ...
Any suggestions?
Best regards & thanks for your valuable help
_atari_
janneman said:
I know - there are also opamp based examples in the WM8740 in the datasheets. Considering costs for decent opamps and other parts - it will be no 'simple' solution.
I hope that a simple discrete high quality approach is availlaible.
Most solutions I have seen are using capacitors for coupling. That is something I try to avoid.
atari, can you remind me whether this discussion is about I/V conversion or analog filtering? In either case, I would advise that using op amp is probably going to be the cheaper solution, and unless you really know what you are doing, the better solution. Actually, for I/V my DAC simply uses the passive resistor approach. That's the easiest possible thing one could do. I have op amps in differential-to-single-ended setup for gain and buffering.
ezkcdude said:
It was just so insightfull to watch you linger along
My aim is to build an good and easy DAC. Therefore the "simple" voltage output of the WM8740 looked very simple to me.
But to get rid of the DC and to make it configurable using either balanced or unbalanced output a tranformer came to my mind.
But after listening to all your remarks and discussion my concerns started to raise. Perhaps a simple capacitor based solution is easier.
Due to the fact that there seem to be no "simple" solution and the very limited availability of the WM8740 (only direct from Wolfson with high shipping costs - see my post for spare parts). I will perhaps switch the implementation to th AD1955 DAC and some I/V conversion after it.
From your remarks I have seen taht my idea using transformers is viable (beside there is still some dispute if a capacitor is used in connection to ground or not.
My aim now is to get hand on some WM8740 and try to prototype a bit. There should be a way to get some cheap (50-80$/pair) transformers to just try it out.
If I cannot get my hands on some WM8740 I will use the AD1955 and start prototyping with that. In the later case I will probably use opamps to do the I/V conversion (@25$ per OPA627 still pricy - but perhaps there are cheaper alternatives).
After all the prototyping is done succesfully I will update the schematics and perhaps get some PCBs.
Thats my result from your discussion.
So my last question is: Do you know of any transformer that is usable in my configuration?
During the next days I will try to figure out some more specs from the data sheets helping me to decide if and which transformer can be used.
janneman said:
Hi, Jan.
I added the cap because someone said that AOUTP and AOUTN each had a 2.5 volt offset. If that's true, each half of the primary winding has a DCR of about 20 ohms and would ideally draw 125mA from each output, which totals up to 1.25 watts for both channels.
That didn't seem like too good an idea to me so I added the cap and suggested it be fairly large to provide a low impedance path to ground for any asymmetry between AOUTP and AOUTN.
That make sense to you?
se
Steve Eddy said:
That sounds quite reasonable. But what will happen if the dac does not play any music. Both channels are on 1/2 VCC then. The capacitor will fill. Fine. After resuming playing you will start playing with the middle of the winding connected to 1/2 VCC ???
It should be ok - as the transformer cares only about AC.....
It seems to work .... did have overlooked something?
_atari_ said:
Well, keep in mind that the DC offset is there all the time when it's on. Both when the DAC is playing music and when it's not playing music. So for all intents and purposes, the offset doesn't change.
Also, under AC (i.e. music) conditions, the outputs only see across the primary. They don't see the center tap. The only time the center tap comes into play is if there is an asymmetry between the positive and negative outputs which should be quite small if the DAC is of any decent quality.
se
_atari_ said:
Well, first, keep in mind that the center tap and therefore the capacitor are effectively out of the circuit as far as music signal goes. When you have equal but opposite signals on each end of the primary, no current flows through the center tap and therefore no current flows through the capacitor.
Second, for what may flow through the center tap, the capacitor acts as a high pass filter, not a low pass filter. So the larger the cap, the lower its cutoff frequency. You want a low impedance path to ground from the center tap and to have that for the lowest audio frequencies, you need a larger cap, not a smaller one.
se
Why not use a transformer without center tap ?
It works, and somebody claims it is even better.
I think some of us need a little more enlightment about transformers.
What I believe to know so far is:
* The transformer just transformes the load impedance on the secondary back to the primary.
It goes with turns ratio ².
So a 4k load on the secondary of a 1:2 transformer will be a 1k load for the source that drives the primary.
Even if the manufacturer declares it as a 150 ohm : 600 ohm transformer.
So, why does the manufacturer say it is a 150 ohm : 600 ohm transformer ?
Is that just a recommendation to have a 600 ohm load resistance ( as a minimum ? ) ?
* the primary shall be 10x the source impedance.
* The windings have a DC resistance. Let's say 35 ohm : 70 ohm
What happens to the primary impedance if the load on the secondary is 40 ohm or a short ?
Is it limited by the primary DC resistance ?
It works, and somebody claims it is even better.
I think some of us need a little more enlightment about transformers.
What I believe to know so far is:
* The transformer just transformes the load impedance on the secondary back to the primary.
It goes with turns ratio ².
So a 4k load on the secondary of a 1:2 transformer will be a 1k load for the source that drives the primary.
Even if the manufacturer declares it as a 150 ohm : 600 ohm transformer.
So, why does the manufacturer say it is a 150 ohm : 600 ohm transformer ?
Is that just a recommendation to have a 600 ohm load resistance ( as a minimum ? ) ?
* the primary shall be 10x the source impedance.
* The windings have a DC resistance. Let's say 35 ohm : 70 ohm
What happens to the primary impedance if the load on the secondary is 40 ohm or a short ?
Is it limited by the primary DC resistance ?
janneman said:
If you tie the center tap directly to ground (as opposed to through a capacitor) you will indeed have DC current flowing through the primary.
If each half of the primary has a DCR of 20 ohms, and the output offset on AOUTP and AOUTN is 2.5 volts each, then there will be (assuming the DAC chip can deliver it) 2.5/20 or 125mA flowing through each half of the primary to ground through the center tap.
Under AC conditions, current will flow through the center tap to ground depending on the asymmetry between AOUTP and AOUTN.
se
juergenk said:wouldn't have a transformer higher distortion and other issues, especially when DC-loaded, than a capacitor?
regards
Well, if the DC is equal on each half of the primary, the fields cancel out and there is no net magnetization of the core.
And a good quality transformer operated under proper conditions can be virtually distortionless.
Finally, I'm not aware of any capacitor that will give you ground isolation.
se
Bernhard said:
Guess it depends how one defines "better."
Using a center tap can improve common-mode noise rejection.
It also provides a path to ground for any asymmetry of the waveform.
Yes. Though keep in mind that the impedance transformation works both ways. Not only is the load impedance reflected back to the primary by the square of the turns ratio, the source impedance driving the primary gets reflected to the secondary by the square of the turns ratio.
That's partly due to old conventions from the telephone industry carrying over into the audio industry and partly due to the fact that output transformers tend to work best with lower impedance loads.
Output transformers don't use as many windings as input transformers and therefore don't have as much primary inductance. It's primary inductance that basically sets the limit for the low frequency response as it's in parallel with the reflected load at the primary.
So let's say you have 100k ohms across the primary. If your primary inductance is say just 1 Henry, then at 20Hz, the source sees a load impedance of 126 ohms||100k ohms or uh... 126 ohms.
Essentially, yes.
se
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.