Hi
I was a litttle bit confused some postings before, topic LM317 >37V output.
In fact, it is no problem to make a 45VDC psu with LM317 using the "normal" circuit.
There are many professional microphone preamps on the market, delivering 48VDC phantom power, produced with a LM317.
The psu will just not be shortage proof!
In the meantime, I populated my board:
Only the LED's are missing. I could not match two chains of four pieces to get 7 to 7.3VDC out of my 20 pieces yellow led's.
One chain was easy, for the second I got values below 7V or more than 7.5V. No combination found.
Franz
I was a litttle bit confused some postings before, topic LM317 >37V output.
In fact, it is no problem to make a 45VDC psu with LM317 using the "normal" circuit.
There are many professional microphone preamps on the market, delivering 48VDC phantom power, produced with a LM317.
The psu will just not be shortage proof!
In the meantime, I populated my board:
An externally hosted image should be here but it was not working when we last tested it.
Only the LED's are missing. I could not match two chains of four pieces to get 7 to 7.3VDC out of my 20 pieces yellow led's.
One chain was easy, for the second I got values below 7V or more than 7.5V. No combination found.
Franz
Last edited:
Maybe useful for other builders:
to select the RIAA caps, there is a fine and not expensive kit available, to measure very exactly such low value capacitance:
You can find this kit in eBay ID 320539668341 or search for "LC Meter" (ready to use version ID 320539668813)
I selected my 14.9nF and in parallel 600pF caps with this gear.
Also a nice tool to measure inductance.
Salas: why not make a LC version for this preamp?
Franz
to select the RIAA caps, there is a fine and not expensive kit available, to measure very exactly such low value capacitance:
You can find this kit in eBay ID 320539668341 or search for "LC Meter" (ready to use version ID 320539668813)
An externally hosted image should be here but it was not working when we last tested it.
I selected my 14.9nF and in parallel 600pF caps with this gear.
Also a nice tool to measure inductance.
Salas: why not make a LC version for this preamp?
Franz
Last edited:
That means about 3mA. Correct region for a degenerated V JFET CCS that will feed them in real life. Maybe you can look for pairs using 2 at a time and a 2k2. Then add those noted as near, in quartets.
I see liberal use of USSR bypass PIOs? K40y-9? In general I haven't seen a build yet with your components style. Will be interesting to know results. What are your Riaa 47n and 15.5n (14n9+0.6n)? Did they measure exact?
If its really 0.5% its a good kit, lacks a case though to be really portable and practical.
I will not make an LC version because of need for very expensive very well shielded coils, still easy to hum, difficult for tight tolerances, plus the need of extra buffers (needs be low impedance LC Riaa not to hum and resonate). Another thing is I have some experiences with very highly regarded LC Riaa tube phono stages that did not play better (to put it mildly).
Yes, and the Leclanché MKP's, lets look what is the result!
I am also looking forward for possible improvements.
I like this euphoric moments
First R12/C9
then
the Salas PSU
Smile
Franz
Its nice to discover it in steps. I agree. The way you proceed, the steps don't seem far away anyhow.
Its about the same thing, but yes why not. Chain 100R at its source, return resistor's free end to gate, attach LEDs from there to ground. Use about 30V across.
Its very tone tunable. But don't try other parts until with its V1.2 partner reg. It will change so much from 317 initial impressions to V1.2 that it will be night and day. How near are your JFET matches per position/channel? You have to test the AC gain in the end, for proper channel matching via tweaking of 1st stage Rs.
I finally managed it to select four led's for the bias. It was a big help to use a 2SK170 / 24VDC on the testbench, as the results where exactly the same like in the final circuit (the voltages with this new testbed where higher than with the 10VDC/1k test).
First, nothing worked :-(
I had a completely wrong connection of the BC550C in my pcb. A layout error.
No problem to bend the pins and insert the transistors correctly.
Then, I thought, I power it up, using the 24VDC testbench psu mentioned above.
As result, the leds lighted up for a short flash, nothing worked well, I measured strange result (the LM7824 testbed-psu DC was climbing from 24VDC to 32VDC...).
O.K., I decided to make a quick and dirty 45VDC LM317 based psu and the thing powered up correctly.
To reach the correct DC balance in the first stage, I had to replace the 4R7 R2 by a 10R resistor. I measure now 11.9V on the BC550 collector.
R6 had to be set to 48.8 Ohm to get 10.5VDC at the drain of the second stage.
The amp is working now in the first testsetup and the result is already very good. Absolutely silent, no hum. But for the Denon DL103R I would like some more gain (70 to 75dB).
The channel matching seems to be fine, but I will measure it later. The FET's are within 0.1mA per stage.
It is interesting, how long it takes after powering up till the DC measurements are getting stable. About five minutes!
Ahh, and I also made the R12/C9 mod before using the preamp.
Thanks for all this good stuff!
Franz
P.S.
Next step: an enclosure. Step after: Salas PSU...
First, nothing worked :-(
I had a completely wrong connection of the BC550C in my pcb. A layout error.
No problem to bend the pins and insert the transistors correctly.
Then, I thought, I power it up, using the 24VDC testbench psu mentioned above.
As result, the leds lighted up for a short flash, nothing worked well, I measured strange result (the LM7824 testbed-psu DC was climbing from 24VDC to 32VDC...).
O.K., I decided to make a quick and dirty 45VDC LM317 based psu and the thing powered up correctly.
To reach the correct DC balance in the first stage, I had to replace the 4R7 R2 by a 10R resistor. I measure now 11.9V on the BC550 collector.
R6 had to be set to 48.8 Ohm to get 10.5VDC at the drain of the second stage.
An externally hosted image should be here but it was not working when we last tested it.
The amp is working now in the first testsetup and the result is already very good. Absolutely silent, no hum. But for the Denon DL103R I would like some more gain (70 to 75dB).
The channel matching seems to be fine, but I will measure it later. The FET's are within 0.1mA per stage.
It is interesting, how long it takes after powering up till the DC measurements are getting stable. About five minutes!
Ahh, and I also made the R12/C9 mod before using the preamp.
Thanks for all this good stuff!
Franz
P.S.
Next step: an enclosure. Step after: Salas PSU...
Last edited:
No hum no hiss, working alright, congratulations! No, not 70-75dB for a DL103R. What is the gain of your preamp, amp, sensitivity of speakers?
The pinch off must be higher in the 170V than in the sim model, I will adjust it so it matches your measured voltages with final source resistors.
You can't get it to work under nominal B+ its normal. Yes the JFETs need some 5 minutes time to thermally settle their bias current. It takes 15mins to listen well too.
The pinch off must be higher in the 170V than in the sim model, I will adjust it so it matches your measured voltages with final source resistors.
You can't get it to work under nominal B+ its normal. Yes the JFETs need some 5 minutes time to thermally settle their bias current. It takes 15mins to listen well too.
Oh, yes I see: I used this gain with my TSD15 before, but it has lower output voltage than the DL103R.
But some dBs more are desireable, not having a big difference to the DAC's output in my chain. But to be honest, I have to measure my new preamp first.
Dont worry, I use a fine balanced active system, high effiency speakers!
Franz
P.S.
Very good sound! Soon, I will do my standard comparison: a vinyl record compared to my musicserver. Recording from the same mastering (Fairground Attraction, First of a million kisses)
Last edited:
Salas
should some other parameters change in your simulation, due to the different 170V model, please tell me!
I will also change the needed parts and again give you the feedback.
Franz
/Edit
I am actually amazed about the sound of Count Basie. This preamp seems to be more "controlled", cleaner than the original Le Pacific.
I am thinking that because of the discrepancy between typical BL model and V model, only adjusting the beta does not cover it, and I will see to get it near to your experimental data first, because I am suspicious it maybe loses gain from target after upping both Rs. I will see if we can get some gain back and how, will let you know. Is it much quiet? I mean is it OK for listening loud enough at say 2hours on your system's volume control?
Without enclosure, surely picks up harmonic noise, plus all other stuff you mentioned, but it looks you are off to a fine start. First of all, use coax for input, its sensitive there. Can you listen decently loud in your system by upping your (sophisticated ) volume control up to a sane point for now?
) volume control up to a sane point for now?