MyRef_C with Ultimate BOM

[troystg]

Bought one of those laser thermometers. I am sure it will come in handy a lot over the years.
Uriah

Yeah to measure the concrete...


"And it's hotter than concrete
In July in Houston
And it'll get worse here
Before it turns nice"

Lyle Lovett - Flyswatter / Ice Water Blues

 

[regiregi22]

Yeah guys, IS VERY IMPORTANT to isolate the LM338 from the case by using an spacer and a washer. Or you can just be sure the heatsink is not going to become in contact with anything. I still haven't tested the PSU in real world conditions (I mean with a Lightspeed), but it should get a bit hot, so heatsinking is neccesary for sure.

I am thinking about what to do with the LDR turning on and off matter. Maybe someone has a solution for that?

 

[bender460]

555 timer?

 

[udailey]

What I was thinking was maybe a 5000uf cap or more on the LDRs after a diode to keep it from dumping backward into the amp. That way the cap supplies a bit of power to the LDRs after the voltage drops to low for the amp to stay on.
Uriah

 

[regiregi22]

Has anybody ever implemented something with the 555?

 

[AndrewT]

see if Forrest Mims booklet is around.
It shows many (28) basic and more sophisticated 555 circuits.

 

[regiregi22]

So you uriah, propose to put a big cap AFTER the regulator, is that right? What is the maximum/minimum/normal consumption of a Lightspeed? If consumption is too low and we need only 2 or 3 seconds "maybe" 5000uF is too much. But is a matter of experimenting...
The bigger ones I have at hand are 2x 470uF pana FM

 

[udailey]

Maximum is 40mA at 5V. Minimum and normal arent to far from each other and are probably around 1mA. I use a far different power supply so I dont think testing mine would show what happens. That why I threw 5000uf out there, simply because it wont be expensive for that cap and it will probably last long enough.
Here is a Nichicon Muse of 10,000uf, small size, good price and will most certainly run for a long time after the amp turns off. If you use this you might consider using the protection diodes in the regulator circuit if you use LM338 or 317.
UFW0J103MHD Nichicon Aluminum Electrolytic Capacitors - Leaded

 

[AndrewT]

I have a 150mF 6V capacitor.
Maybe I should charge that up and see how long the LDRs remain active?

 

[udailey]

Thanks AndrewT!
Uriah

 

[AndrewT]

Hi,
at 1mA of draw for the 4LDRs, 5mF will drop at ~200mV/s
150mF will drop ~400mV/min. I would have to charge it up between every track!

 

[udailey]

So basically any small electrolytic will do the job as long as its not at max or min volume
Thanks Andrew.
Uriah

 

[regiregi22]

Andrew, maybe 150.000uF is a bit too killer for an LDR. Appart that taking in account its tolerance, which easily can be -+20%, we are talking of a variation of -+30mF, which is an absolut value of 60.000uF if I am not wrong. An error tolerance of 60.000uF when we are aiming for 5.000uF or 10.000uF seems too far unexact. Or maybe I am wrong in something. I think you should try what happens when using 5kuF, 10kuF or 20kuF.
But that is just my oppinion

Uriah, I will ask you a favour. Could you add my lightspeed psu schematic the protection diode and the big cap? I am asking becase the regulator already has some protection diodes, and I would like to know where will be placed the new diode.

Just for curiosity, how are you Uriah supplying your Lightspeed? It makes me scratch my head thinking how could be this far different PSU

 

[AndrewT]

Hi Regi,
I was somewhat tongue in cheek suggesting a really big capacitor could be used like a battery to keep the LDRs alive all the time.
The arithmetic tells me the cap needs to be about 10,000 times bigger, giving about 20days usage.

 

[udailey]

Its a lot different. It uses a VVS, Current Pump and VCS. It lets me change impedance at will so I can listen to the differences. Build An Amp if you want to check it out but I dont share the schematic. Redesigning it as I type Fun stuff.
Andrew is showing that we dont need 5k or 10kuf. He tested at 150uf and found the voltage drop was kind of slow. I think that after 4 or 5 seconds it would start to get loud on 150uF. Well, maybe real loud but it shows that if we up the capacitance to maybe 1000uf it should be fine. Should only cost maybe 30 cents.
I know its crazy but I dont have any schematic software. What have you been using Regi? I just draw and then put it on a pcb.

 

[AndrewT]

Andrew is showing that we dont need 5k or 10kuf. He tested at 150uf and found the voltage drop was kind of slow. I think that after 4 or 5 seconds it would start to get loud on 150uF.

No!!!!
0.005F is equal to 5mF is equal to 5000uF is equal to 5000000nF

He calculated from: charge/discharge 1F at 1A and the voltage rises/falls at 1V/s

 

[udailey]

Well, some people use uf and mf interchangeably and I thought thats what Andrew was doing. So whats the arithmetic come up with for maybe a half minute or so Andrew? That way we are okay at regular volume and min or max.
Uriah

 

[AndrewT]

do the arithmetic.

 

[udailey]

Hi,
at 1mA of draw for the 4LDRs, 5mF will drop at ~200mV/s
150mF will drop ~400mV/min. I would have to charge it up between every track!
Do the arithmetic.

I guess I thought thats what you were doing.

 

[regiregi22]

For drawing my schematics I have been using.....Microsoft Paint! No jokes, I like the way to work in it by copy, paste and draw pixel by pixel. If you take ability on it, it can give you some professional results. Here attached is the schematic.

So finally, after all the calculations, what are the final values acceptable?