Salas DCG3 preamp (line & headphone)

diyAudio Chief Moderator
Joined 2002
Paid Member
if you don't approve of source resistors then use emitter resistors in the mirror.
200r on both sides and then a 10r on Q2 side and the 22vr on the Q1 side.
This lets you dial out the offset and lets you see the balance of the LTP.

Big mirror resistors don't harm performance.
Emitter resistors in the mirror I had tried few values. 10-33-47-68-100R. Leaned to third harmonic and the DC turn on bump jumped high. I did not like it.

* They can be seen on the little blue breadboard in crop detail from the first picture of post#2
 

Attachments

  • Degenerators.jpg
    Degenerators.jpg
    84.8 KB · Views: 2,192
diyAudio Chief Moderator
Joined 2002
Paid Member
He, I just found a forgotten one in the camera from the development phase. The mirror BJTs and the the second 9610 on daughter boards. There emitter resistors no more. That pic has much stuff. Slim uPA on main board, LSK170s and various Toshibas scattered around, LSK389 TO-71 lying at front left, 1uF C5 rejected at upper left. A Cosmos Tocos test pot at upper right that Tham will like, DN2540s aside left after cascoded and made not much sense, various C1s and C2s from hands on compensation trials, hey even a current regulator diode under the uPA that was evaluated for stability vs the BF looking at us over the "101" ceramic. I mean by blowing on them with the hot air station "scientifically" evaluated.:D
 

Attachments

  • breadb.jpg
    breadb.jpg
    698.1 KB · Views: 2,163
diyAudio Chief Moderator
Joined 2002
Paid Member
Pottery

What pot to use? 10K-20K-50K-100K? OK not the extremes. Had it with 20K but how is it with 50K that's smoother for turning up and easier on sources? Well its not bad at all. Nice tone, just a bit rounder. Safer. :)
 

Attachments

  • 1.jpg
    1.jpg
    750.9 KB · Views: 2,093
  • 2.jpg
    2.jpg
    656.9 KB · Views: 1,985
  • 3.jpg
    3.jpg
    830.2 KB · Views: 1,936
Andrew,

We are still waiting for you to publish one of your projects/designs.

No need to try to redesign Salas' pre/headphone amp. The distortion plots speak for themselves.

;)
I am following this Thread with interest and was deliberately not making any comment on topology nor performance. It was looking good.
Until Salas said this:
The LTP's tail CCS forces the units to bias at about 2.5mA Id per LTP side.
That is wrong and I felt forced to speak out and correct his error.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Rounder in tone?
Can you describe that. I have only tried 50K. I am pleased.

Like when I was listening to Carey from Joni Mitchell's Blue in 192K FLAC with the 20K pot it was upfront and pristine but with the 50K one it came across bit less tiresome. A small difference but it could make the song passable without reaching for testing another filter from the DAC's menu. In this video the Appalachian dulcimer is sounding much softer in tone than the one in the record which is much tinnier in comparison. It won't make it soft like that, it will just put it in the recording space hanging a bit better. (https://www.youtube.com/watch?v=A1kcTjtdVE8)

Maybe it has to do with the necessarily narrower resulting bandwidth vs 20K. Then again all 20K or 50K pots won't be the same, but at least I had same cheap ones of various values. Its more of a system matching choice thing each builder has to do for himself.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Another Salas' interesting design! Thank you for sharing, it definitely deserves a try.

I have a few questions:

1. uPA68H (old, fat) or uPA68HA (new, slim)? You had the chance to test both, any prefrence?

I put the fat ones in today to answer you better. But I could not find any subjective difference. Maybe its the aural memory fainting between listening with the slims then desoldering them, soldering the others, and let warm up, but nothing striking.
 

Attachments

  • Fatboy.jpg
    Fatboy.jpg
    483.6 KB · Views: 1,009
diyAudio Chief Moderator
Joined 2002
Paid Member
Noise floor pics with the volume pot at max. Even with power cords surrounding, a mains earthing lead to the preamp's chassis can do a decent job of chasing some hum field gremlins though. Long lead and going around through the bench PSU's route to mains is poor practice nonetheless. But at -121dB 50 HZ hum and -132dB 100HZ buzz maybe just passable for an open test build on a bench using no shielded coax signal wiring and a silly simple non battery PSU :D
 

Attachments

  • Overview.jpg
    Overview.jpg
    591.1 KB · Views: 2,732
  • Earthed.jpg
    Earthed.jpg
    745.1 KB · Views: 2,752
  • After-Earth.jpg
    After-Earth.jpg
    545.3 KB · Views: 1,793
  • Before-Earth.jpg
    Before-Earth.jpg
    504.5 KB · Views: 1,863
diyAudio Chief Moderator
Joined 2002
Paid Member
Thank you for the extra time, Salas.

The slim ones were around 8mA Idss when the fat ones are around 6mA Idss (at 10V). The slim ones were closer to one another between channels. The slim ones tend to have less Vgs (off). Given all those differences I can only see very small differentiations in measured THD. The circuit staging of bias and other parameters prevail over small Yfs and Ciss Crss differences at set point.

When you want different enough result try matched K170s. While matched K117s are behaving like the uPAs. LSK389 is interesting because it gives the more Yfs but has different THD trend than the K170s. Has less pF than those. The pads for uPA take one Toshiba looking forward and one looking the other way. Those DGS-SGD pins are printed on the silkscreen for no confusion.
 
Looking forward to this.

I am starting to put some parts together and of course I have a question. It looks like I can use matched 2SK117's in place of the uPA68HA's. As I have a bunch of these it would be great to be able to use them, but if you recommend I buy some uPA68HA's I will do that. Please note that my 2SK117's are GR grade. I tried to figure this out but can not find a data sheet for the uPA68HA's.

Thanks
Jim
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Looking forward to this.

I am starting to put some parts together and of course I have a question. It looks like I can use matched 2SK117's in place of the uPA68HA's. As I have a bunch of these it would be great to be able to use them, but if you recommend I buy some uPA68HA's I will do that. Please note that my 2SK117's are GR grade. I tried to figure this out but can not find a data sheet for the uPA68HA's.

Thanks
Jim
Hi Jim! I know its difficult to find comprehensive uPA data so I had posted its best datasheet in reply #25 :)

2SK117s >3mA idss are usable and alright. GR range covers that also. I had measured with such and they compare very nicely to uPA. When same idss they look close enough on the tracer too. Beta tester had built a pre on matrix board with those (face to face bonded) as well.

News are I had finished with the minor practical stuff so I completed with Crt the little corrections on the protoboards. Added MKP2 C5 frame, corrected the silkscreen orientation of Qa, fixed correctly mirroring pads of VR1s orientations, added alternative SMD1206 pads for Cb Cc, added output bleeder resistors to DCSTB. Now they are V1.02 and sent over to Tea.

I gave my finished Pre, PSU, & I-Select modules to Michael yesterday (the first beta tester that I had to meet in months) since he has some expertise with perspex front panel manipulation and he had originally wanted to box mine up too in the end using a cheap chassis. He uses his matrix build for long time now listening with overhauled Eminent Technology LFT8s and he also made a copy for his friend Chris who listens with vintage IRS Gamma. They usually employ DIY CFA amps and have very good TT rigs etc. I gave him an empty V1.0 proto-board to finish his build nicer looking too. He is seldom in posting here though.

I will write a BOM and few construction tips at a point. Its your job to lobby Tea for a near future GB though because I don't do organizing stuff myself ever. Too attentive a thing for me :D

If you will like it we will possibly make separate channel boards higher rails hotrod version some time too. No worries. :up:
 

Attachments

  • DoneDCG3proto.jpg
    DoneDCG3proto.jpg
    304.4 KB · Views: 1,674
  • I-select+dcstb.jpg
    I-select+dcstb.jpg
    543.6 KB · Views: 1,006
Great news Salas. Seems I was reading too fast to spot the link, I was overly excited. I have been looking at building a pre and almost dug out the old tube collection to search for 6V6's. Will build it some day, but the new pre has just the right gain, drives headphones and I know the design, PCB and Group Buy will all be excellent.

PS:
Sorry but I will wait patiently for TeaBag's GB or at least bear my anxiety quietly. Don't want to distract him from getting things ready!!!

Thanks to you, CRT and TeaBag. A great team.

Jim