I don't know in general use, they remedied some loose barrel contact high pF old scrawny freebie cables for connecting the deck to the pre-amp so I could hear they helped it vs before. Like less noise more heft & depth. It has old type JRC op-amps inside not good for driving capacitive loads. So there were legit contact quality and electrical reasons for that anecdotal improvement. I have Mogami W2497 8mm diameter very low capacitance cable on the ph and dac sources anyway. But the terminals body diameter it calls for do not fit the deck's clearances.
I just thought I tip you off not on subjective grounds but because pro quality good construction interconnects are hard to come buy cheap and even more rarely found in RCA terminated versions. They usually have guitar jacks or XLR.
DCG3 as a flea power speaker amp
Good afternoon Salas,
my question is about the usage of the DCG3 as a speaker amp. I do have a pair of 10 Ohms 100db/W speaker and they work great with a EL84 flea power tube amp. I would like to try them directly on the DCG3. In order to get about 1.5 W I would need to raise Id to about 0,4 A. I think this should be doable with IRF fets if sinked in a proper way. PSU wise this should be ok as well. What do you think about and/or recommed?
Cheers, Ernst
Good afternoon Salas,
my question is about the usage of the DCG3 as a speaker amp. I do have a pair of 10 Ohms 100db/W speaker and they work great with a EL84 flea power tube amp. I would like to try them directly on the DCG3. In order to get about 1.5 W I would need to raise Id to about 0,4 A. I think this should be doable with IRF fets if sinked in a proper way. PSU wise this should be ok as well. What do you think about and/or recommed?
Cheers, Ernst
Hi Salas and all,
I think I'm done with my second DCG3 setup and fine tuning. SQ is amazing.
I use Khozmo Relay now and it is leather type unit with 64 steps plus remote. I also used new Focusrite 2I2 3rd gen.
Please see attached files for reference.
It is input gain difference between my initial test with 2nd gen and old attenuator vs. 3rd gen last test with Khozmo. 3rd get can’t go higher than -7dB, since otherwise it is in saturation.
I think I'm done with my second DCG3 setup and fine tuning. SQ is amazing.
I use Khozmo Relay now and it is leather type unit with 64 steps plus remote. I also used new Focusrite 2I2 3rd gen.
Please see attached files for reference.
It is input gain difference between my initial test with 2nd gen and old attenuator vs. 3rd gen last test with Khozmo. 3rd get can’t go higher than -7dB, since otherwise it is in saturation.
Attachments
Hmm... for 14-15V clean pk-pk to the 10Ω load so to have ~3dBW headroom you would want 0.8A CCS and there is some IXYS with such kind of IDSS to replace the DN2540. That makes it a well cooking Class A amp and the relay type must man up to the job of passing larger current too. For PSU with the proper juice I would recommend two L-Adapter stacked like batteries for virtual common point in this case.
Your old attenuator was obviously compromising things a lot vs your new ladder type. Was prone to gathering interference.
There are some 60Hz components producing harmonic noise to still try clean better I think. Not too high but you already did that a little more successfully in your other build. Why your ARTA charts can't match the THD results of your RMAA reports? Driver software issue?
You could calibrate with gen and scope the software + interface so to have dBV and not dBFS scale which is arbitrary. There is calibration menu procedure in each package. 0dBV=1VRMS when 0dBFS is whatever clipping roof some digital gear has.
I did all what I gained knowledge-wise during my first unit setup in order to reduce that 60Hz with it harmonics noise. It is still there... and BTW it is totally gone when I disconnect attenuator and leave inputs open on DCG3 PCB.
I need to see how to GND Khozmo attenuator properly. Since it’s GND L and R must be shorted and connected to its PSU 0v output. This is one join point of L/R GNDs.
My GND is isolated between L and R till i-Select when I use Apls potentiometer.
Maybe I need to mod my I-Select to separate GNDs too. At that way, I’ll have only one join GNDs point and it is on Khozmo.
Just got my new Focusrite and will calibrate it tomorrow. Thank you for tips.
I need to see how to GND Khozmo attenuator properly. Since it’s GND L and R must be shorted and connected to its PSU 0v output. This is one join point of L/R GNDs.
My GND is isolated between L and R till i-Select when I use Apls potentiometer.
Maybe I need to mod my I-Select to separate GNDs too. At that way, I’ll have only one join GNDs point and it is on Khozmo.
Just got my new Focusrite and will calibrate it tomorrow. Thank you for tips.
I am in the process of wiring my DCG3 and noticed that the headphone jack I am using grounds to the chassis, bring both left and right grounds together. I haven't connected any grounds to chassis yet, so I am wondering if this is as good a place as any to do so.
It might be good to check your headphone jack to see if it grounds to chassis also. Could be causing a ground loop.
Rush
Salas, I bought the full kit from Tea. I just got it together but alas it doesn't work properly. the power supply is okay + 17.3, -17.3 both sides. when i powered up the DCG3 board. led's lit up and relay clicked right side had bias of .2 to .3 volts and offset of 600mv. left side had correct bias but no offset. attached mosfetrs to heatsink with mouser silpads. i took board out of preamp and off of the heatsink- all parts are correct and in right place all solder joints look good and no bridges. i measured from v+ grnd and v- to mosfet tabs when not attached to heatsink or power supply. i got 1246 ohms from M1 tab to v+pad on the right side the side with low low bias.any ideas??? thanks for your help salas!!! mark
About mounted MOSFETS you want to know there is no continuity from tabs to sink screws.
If there is also a dubious one for resistance to tab when off sink, remove it and confirm its status with a transistor tester or replace it straightaway.
When the system is sinked and powered up you may compare Vgs of MOSFETS between channels. Also BJTs Vbe, and DC voltages at various circuit nodes.
If there is also a dubious one for resistance to tab when off sink, remove it and confirm its status with a transistor tester or replace it straightaway.
When the system is sinked and powered up you may compare Vgs of MOSFETS between channels. Also BJTs Vbe, and DC voltages at various circuit nodes.
Hi Rush,
My headphone jack GND pin is insulated from chassis. Its body is connected to chassis, but not its GND. I actually run L/R coaxes to it, but only one coax braid is delivering GND to it. Both channels braids on the same coaxes are connected to GND on DCG3 PCB.
All tests are indicating that I have GND loop, but I can’t find it yet. It is no 60Hz spike at all when I disconnect inputs of my Focusrite or I short inputs
There is a chance that I am picking up some transformer induced flux since I’m using 3 toroids in that setup.
I guess that my next attempt will be dedicated to use of Mu Metal sheets. However, these View attachment 1
are quite pricey and I’m not sure about these effectiveness for my unique case.
Thanks Salas, seems I better make a 2nd one. Maybe I'll come back asking about the CCS.
Cheers, Ernst