i have mount my transistors the exact same way as audiong and with 1/8 thick angles (same as audiong?) and i believe heat transfer is poor...and i m running around 30W, CCS. I d recommend 3/8 thick angles. Since i wanted to find the optimal sweetspot for my system (lowthers) regarding CCS, i tried 3 salas CCS as i m fond of salas stuff. first was salas shunt reg 1.1 CCS, very natural but a bit slow. Then salas shunt 1.2 CCS, less sloppy but a tad bit less natural, and less bright. Last the one posted on page 14 (i think), Unreal fast and dynamics but somewhat bright, but shows a LOT of details. I m not sure which one i ll keep but for sure playing with parts inside each CCS change the sound and the last CCS shows LOTS of potential (maybe it s perfect and the problem is elswhere though). Finally, i also tried a unregulated bias supply and it does sound a bit different: more relaxed and detailed.
Last edited:
good work Audion, i have the same G point as you with my hammond 21V 2amp . but i use fixed bias with LM317 ; for me it sounds better than bypassed source resistor or even more unbypassed one .. ( too much feedbach for my taste) . don't be afraid to try a good EVox or Rifa 100000uf bypass cap with your source resistor 🙂
Hi Antoinel,
Your explanation is probably correct - whatever the reason, the sit amp's response is very sensitive to different speaker loads and also to the source preamp. The good news is that the differences all sound more than acceptable to me and I have yet to encounter any egregiously 'bad' combinations - this is the magic of the sit amp.
Having dabbled in diy audio for the last 20 odd years, I have accumulated quite a stash of parts for my projects. Except for the 2SK82 and Hammond 193v, I did not have to buy any parts for the project. If you had to buy all needed parts, my estimate would be as follows: - equivalent case from Par-Metal - approx $ 150.00 , 2x2SK82 - $ 48.00, 2X Hammond 193V $180.00, Power supply Torroid - Antec AS4220 $51.00, Torroid Case - Antec CA-400 $26.00 Power supply caps,coupling caps,resistors etc. $150.00, Misc. -TO-3 mounting hardware, fasteners, wire etc. $50.00 - Total $655.00
Your explanation is probably correct - whatever the reason, the sit amp's response is very sensitive to different speaker loads and also to the source preamp. The good news is that the differences all sound more than acceptable to me and I have yet to encounter any egregiously 'bad' combinations - this is the magic of the sit amp.
Having dabbled in diy audio for the last 20 odd years, I have accumulated quite a stash of parts for my projects. Except for the 2SK82 and Hammond 193v, I did not have to buy any parts for the project. If you had to buy all needed parts, my estimate would be as follows: - equivalent case from Par-Metal - approx $ 150.00 , 2x2SK82 - $ 48.00, 2X Hammond 193V $180.00, Power supply Torroid - Antec AS4220 $51.00, Torroid Case - Antec CA-400 $26.00 Power supply caps,coupling caps,resistors etc. $150.00, Misc. -TO-3 mounting hardware, fasteners, wire etc. $50.00 - Total $655.00
Hi Juanitox,
Thanks for your kind words. The source resistor (3 ohm) in currently bypassed by over 240,000uF. The little button like cap in the picture is an ELNA 0.22F 5.5v capacitor (yes 220,000uF ! ) and together with the two (in line) much larger caps (10,000uf 10v) totals 240,000uf, not including a 100uF and 10uF cap lurking unseen beneath the circuit board .
As the bias voltage rarely exceeds -5V at the selected operating point, I am comfortable with the ELNA's 5.5V rating.
Before switching to the ELNA's , I used two VERY large 100,000uF 50V Connell Dublier caps. I was initially very dubious about using the ELNAs. How could this tiny midget of a cap provide the huge capacitance needed even at 5.5V ?
Boy was I wrong ! Not only did I not notice any change in gain, the amp sounded smoother and the sound stage depth and width stayed the same, when I replaced the Connell Dubliers with the ELNAs.
I have attached a picture of the little ELNAs so you can have a better look.
Thanks for your kind words. The source resistor (3 ohm) in currently bypassed by over 240,000uF. The little button like cap in the picture is an ELNA 0.22F 5.5v capacitor (yes 220,000uF ! ) and together with the two (in line) much larger caps (10,000uf 10v) totals 240,000uf, not including a 100uF and 10uF cap lurking unseen beneath the circuit board .
As the bias voltage rarely exceeds -5V at the selected operating point, I am comfortable with the ELNA's 5.5V rating.
Before switching to the ELNA's , I used two VERY large 100,000uF 50V Connell Dublier caps. I was initially very dubious about using the ELNAs. How could this tiny midget of a cap provide the huge capacitance needed even at 5.5V ?
Boy was I wrong ! Not only did I not notice any change in gain, the amp sounded smoother and the sound stage depth and width stayed the same, when I replaced the Connell Dubliers with the ELNAs.
I have attached a picture of the little ELNAs so you can have a better look.
Attachments
I think ilimzn mentioned previously that the SIT device is also very sensitive to power supply voltage fluctuations, so it looks all-round vulnerable to voltage and current stability. I suspect it would also therefore be sensitive to heat-induced changes in its operating parameters, though to what extent I could only guess.
I have asked Paul Hynes to develop a shunt regulator for me for this SIT project. Yup, that should be the cat's a**.
I have asked Paul Hynes to develop a shunt regulator for me for this SIT project. Yup, that should be the cat's a**.
Hi Pidesd,
I spent a considerable amount of time and effort in ensuring that the mating surface -angle aluminum to heatsink was as smooth and level as I could make it, starting with a file then sand paper and finally lapping compound. You can see from the pictures I posted that I used five bolts - one at each corner and one in the middle to bolt the angle to the heatsink trapping a layer of thermal 'goop' in the process.
Using 3/8 inch aluminum angle stock instead of 1/8 inch to mount the 2SK82 would provide more thermal heatsinking mass and in that regard would be an improvement, but unless you take the necessary care to ensure a good heat transfer path to the finned heatsink, would only confer a limited benefit.
The results of your CCS efforts are very interesting and I have filed them away for future reference - always wanted to try a CCS, just too cheap to spend the money to buy the necessary parts not already in my stash !
I spent a considerable amount of time and effort in ensuring that the mating surface -angle aluminum to heatsink was as smooth and level as I could make it, starting with a file then sand paper and finally lapping compound. You can see from the pictures I posted that I used five bolts - one at each corner and one in the middle to bolt the angle to the heatsink trapping a layer of thermal 'goop' in the process.
Using 3/8 inch aluminum angle stock instead of 1/8 inch to mount the 2SK82 would provide more thermal heatsinking mass and in that regard would be an improvement, but unless you take the necessary care to ensure a good heat transfer path to the finned heatsink, would only confer a limited benefit.
The results of your CCS efforts are very interesting and I have filed them away for future reference - always wanted to try a CCS, just too cheap to spend the money to buy the necessary parts not already in my stash !
CCS is fine per se , but from Signal Energy Preservation point , choke is better
choke have tad less detailed , but also tad livelier sound
choke have tad less detailed , but also tad livelier sound
I suspect some detail is lost in the choke's wire (how many hundred feet?) with its suboptimal insulation, and in the iron, if any. I personally generally prefer choke loading myself.
Rdc of choke isn't so important , as hysteresis of magnetic field is
there is , also , parasitic capacitance of coil , but it isn't so much of an issue , in this currents/impedances category
Hi AUDIONG.
You are fortunate to own this state of the art [or science] and high performance -low cost amplifier. A keeper for a long future of enjoyable music satisfaction. Lowers blood pressure!
How many amps have you built over the past 20 years? Most importanly are any idle because their technology fell out of grace? Maybe you'd consider selling them. This is a suggestion to DIYers at large to sell their [amplifier] creations which have become in their opinion [their hearing] obsolete. Engrave your name on it, provide a schematic and rake in the dough. Best regards.
+1. The more obsessive among us will be rich.*
* Off-balance sheet debt for new purchases excluded from the calculation.
got any prices or info ,used quite a bit of Pauls stuff so very interested
I don't have any info, Kasumi. I haven't decided on an operating point yet, but once I have, Paul will fashion some regs. He'll have the design details worked out, so any subsequent orders can ride on those coat tails.
No AS-4440's Antek - AS-4440 in stock for a month now and no more possibly to the end of July. I have found another option but the cost was over double the Anteks! Options?
Regards,
Dan 🙁
Regards,
Dan 🙁
Web programming error? Try and place an order and they're out of stock. First response was end of June. Now it's possibly July 23rd.
Regards,
Dan
I have been in touch with John. July 23 is the date he's quoting more toroids to be in.
Regards,
Dan 🙂