Here you go, new group buy starting:
http://www.diyaudio.com/forums/group-buys/216891-paradise-phono-stage-97.html#post3374548
http://www.diyaudio.com/forums/group-buys/216891-paradise-phono-stage-97.html#post3374548
Ricardo,As a matter of fact, using the stock 33.3nF in the riaa filter, the most correct value for the corresponding resistor would be 9550 ohm because 9550 x 0.0333 = 318uS
So your choice of 9k6 corroborates my calculations
Of course I would not stop there and somehow reduce the bass resistor and also increase the trebble cap to 11.4n but I am glad you like the aproximation.
please excuse my ignorance because it is certainly here in this thread but what R's do you propose for 11n/33,36n C's?
Thanks
Sven
As I did say
When using J109's in the PSU you may want to lower R108/208 from 3.3k to 1.8k, which will keep Id at 1mA. But if you leave the 3.3k resistors in place it will still work nicely at an Id of 600uA, it will lower the GBW a tiny bit. The point is you could temporarily use the J109 with the 3.3k resistors, and later replace the J309's with J310's while everything keeps on working
This thing is designed for 1mA so it should run at 1mA, for this purpose a J310 was selected because it fits the job description. You could use a J309 at this point but to have it running at 1mA you need to adjust the resistor. How important is this? I do not know, I'm just de designer of the PSU, these are the values I selected, calculated, simulated and tested, and it works great. Would other values work, maybe, but I should have to re-calculate, re-simulate and re-test. The point is, when designing these things you make decisions and now that it is build time you stick to them, otherwise the whole design process is dismissed, and you have to start all over again.
So what should you do? get a few J310's
and make it work.
Hi Sven
There are no "perfect" values for resistors when using 11n for the high freq cap and 33.3n for the mid freq turnover.
IMO the relation between the 33.3nF cap and it's series resistor is fundamental so to get 318uS... As I said 9550r x 0.0333 = 318uS
When using that combination, the 11n cap should be raised to 11.4nF to cope with Stanley Lipshitz rules.
If you increase the 11nF cap to 11,4nF you might loose some detail and "air" so it is necessary to reduce the "bass resistor 73.5k" accordingly.
In a first attempt I would use 143k parallel with 140k to reach 70742ohms.
But as we all know, rules are only rules and no rule is good if it can not withstand some exceptions.... The preconized Paradise values provide a "house curve" that is quite agreable with loads of bass slam and pinpoint detail. In my system it was a little "off shore" so I recalculated it.
Just to remenber, I am actually using the following combination:
147k // 147k = 73700 ohm for the "bass resistor"
32.1nF in series with 9902r for the 318uS middle point
11nF for the trebble resistor
Last edited:
I assume J309 is supposed to be J109 in your post. In this case it would make sense to add your comment regarding R108/208 to the assembly guide/BOM. Everybody who participated in the parts GB recieved J109 and it was news to me that I should (may?) change resistors.
Ricardo: Thanks. Reason for my question is I would like to avoid any additional emphasis of anything since my system has plenty resolution anyways. However, now I have 11n/33,36n caps which I ordered specifically for the paradise. I'm a bit reluctant to change these, changing a couple of R's however is ok.
Since it will probably be lost I'd like to ask my other question again: is it ok to mix SK170 and J109?
Regards
Sven
Last edited:
No and Yes, sorry for that. I did say J309 and the resistor change is for using a J309. I misread the post, now I did check the effect of using a J109 and I have to conclude that the existing 3.3k resistor is o.k. (Id when using the J109 will be about 1.2mA).
And I overlooked that
Not needed, see J309/J109 explanation above.
Last edited:
Argh... we certainly managed to confuse everybody now.
Yes I did
To restate and be totaly clear:
The J109 can be used as a direct replacement for the J310
Hi SvenRicardo: Thanks. Reason for my question is I would like to avoid any additional emphasis of anything since my system has plenty resolution anyways. However, now I have 11n/33,36n caps which I ordered specifically for the paradise. I'm a bit reluctant to change these, changing a couple of R's however is ok.
If you must use those caps, and you want a linear response, than I suggest:
9532ohm instead of 9k1 i series with 33.36nF cap
If even after that you feel there is too much trebble enfasis, add 400pF styroflex to the 11nF cap.
If you feel there is too much bass enfasis, use 140k // 143k to get 70.7kr on the bass resistor instead of the stock 73.5k
Please note that these are simulated alues to get a straight response... these are not my prefered values but IMO they will work fine.
Success!
Well, partially. Regulators are working. Rails are a bit to high, 27.2V so the heat sinks get too hot, 60°C after 30 minutes. Will probably have to use larger heat sinks or go with the Preregulator...
Well, partially
. Regulators are working. Rails are a bit to high, 27.2V so the heat sinks get too hot, 60°C after 30 minutes. Will probably have to use larger heat sinks or go with the Preregulator...An externally hosted image should be here but it was not working when we last tested it.
Success!
Well, partially
An externally hosted image should be here but it was not working when we last tested it.
You could split the PSU capacitor's in groups of two (so to see) and add a resistor inbetween (5.6 Ohm will be spot on).
Yep will do, give it a bit of air. Do not mount against the board and not to near the capacitors. The power disipation will be about 1 Watt each.