Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

[moved] 48v phantom power switching ( multichannel )

Status
Not open for further replies.

benbiles

Member level 4
Joined
Jan 24, 2014
Messages
76
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Activity points
2,146
Hi,

I would like to switch 8 channels of phantom power supply +48v ( 100 ma approx ) for an 8 channel digitally controlled preamp i have made.

I have 5v logic on/off for each channel to use as switch control.

I've looked at 8 channel darlington transistor arrays ( high & low side switching ) but can't see a way to use them as the phantom power point of injection on the input channels of the preamps is made of 6.8k resistor deviders accross the audio channel inputs & ultimatley share ground that the 48v & pre amp power supply uses.

So switching ground on / off will not work ?

unless i'm wrong here and ther is a way to use the darlington setup ?

I'm asuming mosfet would lead to a similar problem.

I don't want to use relays, big heavy and unessasery for switching 48v 10ma of power on/off per channel at most?

or maybe there are mini relays that are designed for this?

Is there an 8 channel SPST switch that would be a good choice for something like this? would that work?

Or am i missing an obvious trick here with the darlington array ?
 

A PNP transistor turns on a load in the collector leg. The load is referenced to 0V ground. This is compatible with your other equipment.

However your 5V signal cannot turn off the PNP directly... Therefore add an NPN as a middleman. The correct bias V is automatically applied to the PNP.

6942764700_1464926843.png


This schematic corrects the one I posted less than an hour ago.

The 100k resistors are a good idea, to ensure the transistors turn off completely when no bias is applied.
 
Last edited:

Hi,

Other possible solutions:
* use a dedicated high side switch rated for the voltage
* use a photomos relay...benefit: isolation between audio GND and microcontroller GND
* use a reed relay, or other signal relay, maybe bi stable

Klaus
 

Just thought of another problem.

The switch should conenct output to 0v via a resistor after being in on state in order to empty stored DC in 48v blocking capacitors on preamp IC input side.

maybe i'm looking for a 2 pole 8 channel SPI controlled switch array.. thanks for ideas, i'll lool up photomos relay etc..
 

Hi,

Look at the specifications of the ULNxxxx transistor array devices.

Klaus
 

I already looked at the ULNxxx darlington transistor arrays..

They are all common + ? meaning they switch on power by connecting to ground.

I can't use this since I need to switch + and use a common ground.

I think I am looking for an 8 channel common ground darlington array?

This would still not allow connetion to resistor / GND for capacitor discharge in off position.

I think UDN2981 is a high sided common ground darlington array.
 
Last edited:

You are trying to design a high-side or "source" driver. UDN2981 or 2984 is a single chip solution for the problem. A discrete solution would use high-side PNP respectively PMOS switches and a level shifter circuit to control it by ground referenced 5V logic. E.g. a NPN transistor and a few resistors.
 

Hi,

The are all high side darlington arrays ?
most of them are NPN type (Maybe all? I don´t know). NPN usually is used for low side switching.

***
You now want to switch to GND. Therefore you need something like SPDT, but you talk of SPST.
Isn´t is pulldown resistor and a SPST sufficient?

***
I think UDN2981 is a high sided darlington driver that would work ?
Looks good.

Klaus
 

You are trying to design a high-side or "source" driver. UDN2981 or 2984 is a single chip solution for the problem. A discrete solution would use high-side PNP respectively PMOS switches and a level shifter circuit to control it by ground referenced 5V logic. E.g. a NPN transistor and a few resistors.

Yes, discrete solution would involve too much board space, I have 8 channels to switch. I can see the PNP would have a very high switch state and need a NPN in front of it for 5v logic level.


I need an 8 channel logic level controlled switch that flips between ( position ON ) supply +48 or ( position off ) short to GND via a resistor. the UDN2981 presumabley leaves the input floating in off position.

- - - Updated - - -

***
You now want to switch to GND. Therefore you need something like SPDT, but you talk of SPST.
Isn´t is pulldown resistor and a SPST sufficient?

***

Yes I think I need an 8 channel logic controlled SPDT switch array that can handle 50v 10 - 20ma per channel. That way I can short blocking caps to GND via resistor in off position individually for each 8 channels. Do you know if an array like this exists ?

- - - Updated - - -

Yes , sorry Klaus , I moved the goal posts as I was thinking this through !! I went from SPST to SPDT,
Its not absolutly essential but would advantagous to empty phantom power from bloacking caps in the one switch.
 

There are no HV SPDT switches. Use UDN2981 or similar with pull-down resistor.
 

OK thanks FvM,

I'll order a UDN2981 and try a 1 meg pulldown resistor for blocking capacitor discharge and test it switching the 8 channels of +48v from 5v logic. It seams high side multichannel darlington arrays are quite rare ! know of any others?
 

2981 is specified with 200 µA maximum leakage current. Even if we assume that actual value is more likely around a few µA, 1 Mohm is probably too high. I would use a value that is acceptable in terms of power dissipation and quiescent current, e.g. 10 to 50 kohm.
 

2981 is specified with 200 µA maximum leakage current. Even if we assume that actual value is more likely around a few µA, 1 Mohm is probably too high. I would use a value that is acceptable in terms of power dissipation and quiescent current, e.g. 10 to 50 kohm.

OK great, I will test values based on drain speed / quiescent current..

On a side note, I'm using incorect blocking capacitors for phantom power, 33uf 50v where they should be 33uf 63v.
11v DC is getting past the caps ! Can you recomend s good capacitor type to use in this scenario ?

If its a complicated answer, I'll post a new thread ..

I'll try and post back here and give update on UDN2981LW + resistors for phantom power switching.
 

OK great, I will test values based on drain speed / quiescent current..

I'll try and post back here and give update on UDN2981LW + resistors for phantom power switching.

OK, the UDN2981LW is working for the 48v switching. I used at 10k resistor which discharges the 63wv 47uf caps fairly fast.

I'm getting a low frequency tone that ramps down in pitch until completely disapears within 2 second when i switch on the 48v.

I think its comming from the ripple / 0.1uf filter cap I have on the power supply output that is filtering 200 - 400 khz noise , I'll have to play around with this more. maybe its also comming from the blocking caps filling up with charge.

I can digitally ramp down the gain to 0db in software for 2 seconds but the audio is still there. There seams to be no digital mute function on the the pga2500's.

hopfuly I don't need to add a SPDT switch on every audio output channel just avoid this noise !
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top