With the input grounded wont Vds for either device be only 24V? as the output should be 0V in this case?
With 2Vrms out (5.65Vpp) Vds will vary between approx. 21.2V to 26.2V, or am I wrong 😕 ?
Of course if one jFet fails Vds will be 48V.
Could I use lower IDSS jFets? With a 6-7mA jFet the dissipation will lower.
and what if a HV HF transient comes along?
Or a big DC pulse from switching some gear on upstream?
Or a big DC pulse from switching some gear on upstream?
They are too near to each other for mounting sinks, aren't they? 6-7mA will get them down to circa 200mW each (the Idss at 24V is more than at 10V datasheet). Thermal equilibrium will be naughtier for offset, they were getting coarser at 15V to my ears too. There are some drain to gate leakage paths for those devices when used higher if I remember correctly. If you must use them like that, yes they will work, its not optimum, can be risky, to share the regs with another system is not optimum again, but its what you wanna do in the end. Is there a capacitor coupling somewhere in the chain after it? I still recommend you better avoid non optimum use if you can.
Ok, I got your point.
I was looking on several Borbley schematics where he used similar followers with +-24V, however, all circuits have NFB and even DC servo; so I guess those solves the situations you mentioned.
What about lowering the rail voltages to +-20V, use of 6-7mA idss devices and use of heatsinks?
later edit:
My JBoz like preamp would ideally need +-32V so +-20 is already a big compromise that might not work, I have to calculate. I will listen to B1 with +-10 and +-20, but probably another regulator for the preamp will be needed :|
Thanks!
Last edited:
The sound quality will still impress if with no prior experience of it, if you use it high and share reg, but if you use them as I recommend, there is more. I.e. +/- 10V, MKP film cap instead of 100uF lytic over strings of 5 leds, 150mA-200mA current, Mosfets mounted on the floor to sink, different regs for the other stage.
Then again, do you need the the gain of the JBOZ like? Maybe your amp's gain and speakers are capable for SPL with just DCB1, and you avoid all changes, with a transparency gain of shortening the chain?
Then again, do you need the the gain of the JBOZ like? Maybe your amp's gain and speakers are capable for SPL with just DCB1, and you avoid all changes, with a transparency gain of shortening the chain?
No, I don't need that much gain. I recalculated it to my needs to have 23db gain. I plan to use this preamp after a passive I/V Dac End2. I need the -24V because I planned to make it DC coupled by adding a source follower J74 second stage after the first one. Based on first stage Id I calculated the second stage values to have 0V out (+-5mV).
With +-20V I might not get the gain with the THD I plan (~ 0.2%), but probably 17db.
After this an 4x input selector + R2R with Zout 10K follows. Here I need DCB1 🙂
I couldn't source the recommended (BOM) diodes so I used some MUR860's I had leftover from an amp project. Don't know too much about CCS yet. I just followed the instructions with two 68's in place. I do have two 5W 10 Ohm resistors ready to go in later when I learn more about hotrodding this buffer. Should I be concerned about equalizing the +ve and -Ve supplies?
Regards,
Dan 🙂
Last edited:
VZS:
The thing with THD isn't only favorable or not harmonic shift, it cloaks transparency in my experience. Its a measure of linearity more than a measure of harmonic content IMHO. What happens if you make that load resistor on the 2SK170V a CCS?
The thing with THD isn't only favorable or not harmonic shift, it cloaks transparency in my experience. Its a measure of linearity more than a measure of harmonic content IMHO. What happens if you make that load resistor on the 2SK170V a CCS?
According to the schematics it has to sink 20mA (36V/1.8k), so a 20mA CC-Sink will probably work, however, the Zout of this circuit is given by this Rd and CCS's have high Zout. If I use another source follower jFet with high Zin it might work, but I'm not sure 😕
Why do you ask?
Just hooked it up to my Aleph 5 (pics later) and I'm very impressed. First impression is that it's very clear and well balanced. Will have to let my Aleph really warm up before a true test.
Current? Give me a quick hint and I'll check. Again, all I know for the moment is I built it to the specs listed in v4.3 BOM. After a quick review of the this thread I thought the CCS current was set by R1 in your schematic. Thus the reason for the two 68R's in parallel. Correct?
Regards,
Dan 🙂
Since you got ample sinking in the way you mounted the Mosfets on a thick metal box, you may rev up the CCS current by using a combination for getting R1 down to 11R 3W. If you check voltage across R1 divided by its net Ohmic value you know your now current. Will be nice to mod and compare after you listen to it as it is for a while. The low current is for those who don't have space for sinks and/or want the consumption low. The quality is very nice as standard 68//68R. But picks up noticeably if you push the CCS.
Slipped in the 10R's in and everything appears normal. DC offsets are the same. Voltage drops. Does 2.38VDC and 2.04VDC sound reasonable?
Regards,
Dan 😀
Last edited:
almost done
Here's my build so far...
Just need to wire up a power switch (to the front or back panel).
hmm, led strings are not quite what was recommended (1.8v drop each).
Other than that, I'm really happy with the build. Most difficult part was drilling out a square hole for that IEC inlet.
Next build I'll hopefully find something that is pre-drilled.
Here's my build so far...
Just need to wire up a power switch (to the front or back panel).
hmm, led strings are not quite what was recommended (1.8v drop each).
Other than that, I'm really happy with the build. Most difficult part was drilling out a square hole for that IEC inlet.
Next build I'll hopefully find something that is pre-drilled.
Last edited:
Yes its in the region. Depends on individual mosfets Vgs. If you keep it in the end, find a 12R for the 2.38V side so to match the other for current.