L-Adapter

Thanks Salas!
I do not have a scope but the DMM tells me that there is an AC component at the output - 0.010vac approx - unloaded.
When loaded w/ a 3ohm resistor - draw approx 1.6 amps - the output voltage increases to 5.18 vdc and the AC volltage to 0.070 vac! This is not producing the same load as the RPI but is the lowest value i have with a suitable power rating.

This is measured at the actual output terminals so there is no effect from output cables.
The AC input - measured at the input terminals is 9.9vac - so I don't think the V-in/out differential should be a problem.
BTW - by bu HDD i meant a drive i use to back up the library which is powered from the USB port - usually located on my laptop. I have since replaced the externally powered HDD and am using the standard SMPS and it works fine. If not as enjoyably as with the L-Adp!
Thanks!
 
diyAudio Chief Moderator
Joined 2002
Paid Member
What value reservoir capacitors you have on the L-A? Is your trafo 9.9VAC when the DC output is loaded with that 3 Ohm resistor? What was the combined load for RPi+drive? Roughly. Did the sink went too hot? Let me know so I can try analyze for AC ripple and chances of there might be a broken component.
By the way, if there was significant drop across the output cables with the added backup drive, that could have been why the system halted. You should revisit and measure end of line DC in that configuration.
 
Thanks Salas!

The caps are :
10,000
10,000
470
470


I will check the trafo output loaded tonight when I get a chance.
Load? RPI4 - Approx 3 A intermittently. Allo Boss DAC - 100ma?
HDD - 0.75 - 1.0 A intermittent? This is only a guess - nobody seems to be giving this info away.


FYI - I did run the system w/ the USB powered HDD and a 2.5A smps - and the power led on the RPI flashed occassionally - meaning it was dropping below operating voltage. This stopped when I replaced the HDD with the externally powered HDD. So it seems quite likely that it also overloaded the L-A.


Thanks!
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Alright, lets assume 4A peak. That also means 0.4V peak drop across 0.1 Ohm rail cable. Replace the 470uF output capacitor with a smaller 100uF to 220uF. Set the reg at 5V output unloaded. Load the reg with that 3 Ohm resistor and let me know the input AC voltage, the DC voltage across D11, the Vout, the DCmV drop on R2 and the Vbe DCV of Q1 Q2. Check the heat on the sinks.
 
Member
Joined 2004
Paid Member
Depends of the dissipation ( P= (Vin-Vout)*Iout ), and the heatsink's thermal resistance.

The NJW0302G inner thermal resistance (between junction and case) 0.83 C/W, so -for example at 50W dissipation- the case 41.5 C degree cooler than junction.

Typical junction temperature limit is 150 C, so if the case below 100C, yet -it this case- didn't exceed the limit.

p.s.
BTW if you read the transistor datasheet, there is this figure:
 

Attachments

  • Power derating.jpg
    Power derating.jpg
    48 KB · Views: 450
Last edited:
Good to know.
Thanks, Salas!

I haven't put mine in a fancy box yet. But I did hook it up to my NUC media source and was taken by how much more music was available. I've never experienced such a holographic presentation before. It clearly beats my RPi/DigioneSig/Shanti setup.
Gearing up to replace the already decent linear supply on my DAC next. The cost/performance demonstrated on my NUC made trying it in other places a no-brainer.
 
Hi Salas,

I wanted to use a couple of the L-Adapters but have a couple of questions first. I have a transformer with 2 x 12Vac 1.25A secondaries. If joined the secondaries would this give me a 12VAC 2.5A output into the L-Adapter and would it then be suitable for powering a Pi without needing a larger than say 38mm heatsinks?

The other L-Adapter I wanted to feed in to a Sjostrom SSR01 12vdc super regulator. I would need approx 18vdc out from the L-Adapter to feed the SSR01, would this be reasonable to get from a 16VAC secondary in to the L-Adapter?

Thanks
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Hi,

First case should work. The 38mm sink should find time to cool down since computers usually draw high current only in bursts.

Second case is borderline. Dependent on the transformer's regulation percentage and load current asked. You should check if at least 2.5V DC remains across the D11 diode when on the actual application.