# Sjöström Audio DIY forum

## Projects => Power supplies => Topic started by: peranders on March 16, 2010, 05:00:39 PM

Title: JSR03 at 11 V using a LM329
Post by: peranders on March 16, 2010, 05:00:39 PM
Here is a suggestion for achieving 11 V out together with a LM329:
Not mounted:
T1, T3, T5
R4, R5, R6, R12
C9

R1 = 1k0
R2 = 3k9
R3 = 2k2
R7 = 470R
R8 = 1k0
R9 = 1k8
R10 = jumper (wire from a resistor)
R11 = 47R
R13 = 2k2 to 5k6
R14 = 220R (100R may be a bit hard for T6 but should work)

These calculations gives 10.8 V out but can differ since the tolerance of the LM329 is 5%.
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 16, 2010, 07:54:28 PM
Thank you!

Are these calculations for use wit a blue or red LED?

Branko
Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 16, 2010, 09:21:32 PM
Red LED, altough I had in mind the blue but when I came to think about it you must increase the preregulator voltage to perhaps 5 volts. R2 = 3k9 See my edited first post.
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 16, 2010, 10:10:51 PM
Hi

yust to make things clear. You say:

Here is a suggestion for achieving 11 V out together with a LM329:
Not mounted:
T1, T3, T5
R4, R5, R6, R12
C9

Does that mean T2 T4 T6 should be used instead or no T1 to T6 at all? If so could I use T1 T3 T5 (because i have these)

thx

BRanko
Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 16, 2010, 10:32:45 PM
Yes, only one transistor at each position. See the schematics.
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 16, 2010, 10:58:06 PM
Hi,

is it ok to use a bc860b instead a bc860c

Thx

Branko
Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 17, 2010, 06:12:37 AM
Yes.
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 18, 2010, 10:27:29 AM
Hi,

build it yesterday and a short test gave me 10,8 V :)

Is there any way to get 1400 to 1500 mA out of one JSR03 (cooling?)

thx

Branko

Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 18, 2010, 07:48:20 PM
It might work but this is the limit for the LM317. You might also increase the current through R14.
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 20, 2010, 05:40:53 PM
Hi,

today I tried to use the regulator for the first time.

Without load the output voltage is 10,8 V, but with load it drops to 3,8 V (also if I connect the load (DAC 1400 mA) the current at the pos/neg out of the rfb02 drops from 25 to 16 V with a 2x 18 V toroidal transformer in front). The cooling pad for the LM317 reaches 70 C.

What now (I did not increase the current through R14 yet).

Branko
Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 20, 2010, 07:22:39 PM
How much smoothing caps do you have? What about the VA rating of the transformer? If you plan to take out 1.5 A the transformer must have 2.5-3.0 A as rating.
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 20, 2010, 07:25:26 PM
Hi,

at the moment there is no smoothing cap (have one with 4700 uF waiting). the VA rating of the transformer is 2 x 18 V 2 x 1,39 A

thx

Branko
Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 20, 2010, 07:39:27 PM
You must of course have smoothing.
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 20, 2010, 07:47:37 PM
Hi,

OK i'll try now with the 4700 uF Elko i have. Is the type or quality 0f cap important (elko, polypropylene ...)? should it be near by rfb02 or the SJR03?

thx

Branko
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 20, 2010, 08:21:55 PM
Hi,

I've put in the 4700 uF smoothing cap and the voltage with load is now 4,2 V without load 10,8 V (without smoothing 3,8 V an 10,8 V)

What to do?

thx

Branko
Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 20, 2010, 10:19:10 PM
You should try smaller loads to see when you will get a voltage drop.
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 20, 2010, 11:38:01 PM
I dont know how to measure loads. but it seems the voltage drops if the load greater than 800 mA (i have 4 loads in the dac 100+100+600+600; if i disconnect the one 600 it is 10,8 V again). how could i test loads. any way to increase this?

thx

Branko
Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 20, 2010, 11:50:49 PM
Do you have at least 16 V in and this is the ripple voltage excluded?
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 20, 2010, 11:53:17 PM
i have 22 V at the leads of the 4700 uF smoothing cap. there is 18 V in to the rfb02.

Branko
Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 21, 2010, 12:22:43 AM
Check the voltage drop over T3 or T4. It should be stable 2.5 V.
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 21, 2010, 12:31:14 AM
"Check the voltage drop over T3 or T4. It should be stable 2.5 V."

Sorry for this lot of questions. but how to do this?

Branko
Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 21, 2010, 12:47:23 AM
Measure the voltage across collector and emitter.
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 21, 2010, 01:01:36 AM
The Voltages are 5,63 V (Adj. 10,88; Out. 17,19; In 11,56)

Branko
Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 21, 2010, 09:53:58 AM
This isn't over the T3(T4).
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 21, 2010, 01:26:17 PM
Hi,

i measured this over T3 (BD139 STM) every leg aginst gnd.

Branko
Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 21, 2010, 01:47:22 PM
So you have 11.56-10.88 = 0.68 V across the collector-emiiter. It's too little. It should be 2.5 V if the LM317 is working properly. If you really have 1.4 A it's very near the limit for the LM317. May I ask what your load is? 1.4 A for a DAC seems to be a very high current.

To solve your problem you could try to exclude the LM317. Remove C3 and R2, short pin 2 and 3 on the LM317.
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 21, 2010, 01:55:45 PM
Hi,

at the moment of the measurment (I checked right now again) the load was ca. 700 mA. (the dac use is a DDDAC  MKII (www.dddac.de) which uses 24 1543 chips in parallel. each module with 12 chips needs ca. 600 mA and the input board needs ca. 100 mA. If iuse booth dac-modules the voltage drops to 4,2V. with 1 module and the input board it needs ca. 700 mA and the voltage does not drop. it stays at 10,8 V)

i am already finished a second jsr03 with the same values and will measure it in a few minutes.

Branko
Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 21, 2010, 02:06:50 PM
You should try to divide the power distribution into several regulators, or as I mentioned exclude the LM317.

I wonder also what good does it do to connect so many DAC's in parallel?
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 21, 2010, 02:11:19 PM
As i have 4 pcb and parts. what would be better in terms of stability and sound: several regulators or exclude the lm 317?

thx
Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 21, 2010, 05:47:33 PM
This is a rather "massive" design with many audiofilical elements so I recommend that you take this with the designer. When I see the power supply you already have you can with a good conscious exclude the LM317 at least.
Title: Re: JSR03 at 11 V using a LM329
Post by: Branko on March 21, 2010, 07:34:57 PM
hi,

i have further problems with load. i have excluded the lm317 with the same result.

if i drive one jsr03 with more than 500 mA the voltage drops (with or without lm317 build as you recommenden in ten first post of this thread). so even if i try to use two jsr03 i cn not drive a load of ca. 1400 mA? Any chance to solve this problem. i an ready to resolder the things (it is possible to use Voltage from 12 V to 10,5 V and i can use lm320 05 lm431 which you recommend even led color is not important). Do you have a proven design to drive 2 x 700 mA with 2 jsr03?

Branko
Title: Re: JSR03 at 11 V using a LM329
Post by: peranders on March 22, 2010, 10:35:26 AM
When you have heavier loads you must adjust the design and right now approx. 1 A works good together with a LM317 AND a red LED. The limit for the design as it is right now is the available base current for the pass transtransistor. You could also change the pass element to a mosfet but this is nothing I'll support. In order to achieve ultra high performance there are obsticals on the way to a good result.