Using the HYPNOTIZE as a general shunt reg PCB

Status
Not open for further replies.
Jump to Latest
#383
It's also like this schematic, the Salas shunt V1.0 ?
I managed to keep the powerleads at 10cm, but read that it can still benefit from remote sensing.
 

Attachments

  • shunt24vnew.jpg
    shunt24vnew.jpg
    114 KB · Views: 381
#385
shunt PS PCB - trace widths

Hey Salas and hey all,

I haven't been here in a while... As I am catching up with the discussions, I have a question for you guys. You might remember me designing some PCBs for this, which are now in final stage of prototyping.

Regarding this - and I guess the question is mainly for Salas -, I've kept the traces widths at 1.5mm on all main current draw traces (allowing, according to some online calculators for 3.2A through, plenty for most purposes), while the current sense parts of the circuit use 1mm traces (2.4A).

Everything good with my thinking? Any thoughts on particular traces (connections) that would have unseemingly high current through? For instance, any danger in the two different trace widths to throw off the current/remote sensing device?

Thank you very much for your input.

Best,
Radu.
 
#387
Rax said:
I've kept the traces widths at 1.5mm on all main current draw traces (allowing, according to some online calculators for 3.2A through, plenty for most purposes), while the current sense parts of the circuit use 1mm traces (2.4A).
Click to expand...
I don't believe it is current rating that determines trace width.
I think trace resistance and trace voltage drop are far more important in determining trace width.
We are looking at uV values in measuring voltages and sending error signals around this PCB. I Think you should consider uV not Amperes
 
#388
Salas said:
The RS lines sense voltage and are high impedance. Their gauge has no consequence.
Click to expand...
look at post383.

The sense circuit is isolated in the middle of that schematic.
As drawn, one can see the resemblance to a Wheatstone Bridge.

The Wheatsone bridge relies on very accurate balance of uohms to give the error signal at it's output.
I believe that layout is crucial to this balancing of uohms. Trace width may be inconsequential, but traces, their routes and lengths and where the currents travel is very important to getting best performance from the "high impedance" sensor.
BTW,
10mA through the sense circuit when controlling a 10V output is equivalent to a 1k0 impedance. That to me is a low impedance circuit.
 
#389
In my view, when with the 1miliOhm Zo, 1000000x is relatively high so to consider equalizing gauges. Layout is another thing than gauge, as you said. Its primarily named a Kelvin or Thomson bridge what we do here I believe BTW.
 
#398
difool said:
15-28V adjustable, 3A or 6A (Field Coil Driver)
Click to expand...
the 3A/6A option can be switchable or pluggable.
The 15V to 28V adjustable is easy. The variable dissipation that it introduces becomes a problem at high currents, particularly the 6A option.

You need to look at worst case operating conditions to determine what range of powers the two main transistors must be able to dissipate. I fear quite large heatsinks.


If the load were constant then shunt can be made small and cheap and perform very well. Non constant load results in a much more demanding specification.
 
#400
If im using the hypno as intended (as a dcb1 buffer) and adding a few things to the load that need the same voltage (balanced to unbalanced opamp, and active volume control) Do i just adjust the current resistors?

Also need a 5.5v supply for the analog section of a dac chip (has a 5v regulator on board). Should i tape the relay supply with a lm317 style to knock the voltage down to 5.5v, or can i unequally yoke the rails of the b1 buffer and tap one rail with a lm317 for even quieter opperation?

I need to feed this guy.
Linear Regulators - Single Channel LDO - REG102-5 - TI.com
 
Status
Not open for further replies.
Top Bottom