Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Generate voltage pulses for signaling

Status
Not open for further replies.

FireExpert

New Member
I have a need to pulse a DC voltage. The source power supply is 24VDC. The circuit needs to handle 2.5A. The circuit should then be passed through a voltage converter to increase to 40VDC. The desired voltage of the pulses is 32VDC. The pulses need to be controlled with at least 1ms accuracy.

My question is how should I go about generating the pulses? I see a few options:
1. Have two voltage converters - one outputs 40VDC and the other 32VDC. Use a switch to flip between them as appropriate.
2. Use a single voltage converter that can vary the output voltage from 40VDC to 32VDC, or the reverse, within 1ms.
3. Use some other magic method I haven't described.

Ideas?
 
What supplies do you have now, if any?
What's your budget?
 
To clarify, the steady state output voltage is 40 V, and dropping to 32 V constitutes a data pulse? All at 2.5 A?

ak
 
All I have at the moment is the 24vdc. Budget is...how much do I need? I would hope I can do this for under $20. And yes, dropping to 32v,and back, is the data pulse. The circuit needs to handle a max of 2.5A.It may run as little as 100mA - it depends on how much I connect.
 
I vote for boost to 40 V with one circuit, then modulate down to 32 V with something else. What percentage of the time would the output be down at 32 V? This affects the heat, which affects the parts.

ak
 
What percentage of the time would the output be down at 32 V? This affects the heat, which affects the parts.
To expand on that, are you sending occasional control pulses, a string of data, or something else?

ak
 
Uh...yes? The pulses are for 8-bit strings. There will be constant signaling - during active polling the longest "rest" period would be 10ms.
 
So the worst case power dissipation in the modulator is 20 W, and the average probably is around 10 W. Depending on the ambient temperature and the enclosure, you might need a small fan.

You will need over 5 A at 24 V to produce 2.5 A at 40 V. Linear Tech makes the LT1070 and LT1270 series of highly integrated boost converter devices.

ak
 
I understand using a boost converter to get my base 40V from my source 24V - but again my original question...how do I generate the 8V dip pulses? Are you suggesting altering the output voltage of the boost converter?
 
Altering the converter output would probably be too slow. Purely because of the 100mA to 2.5A requirement, your simplest method might be to switch between two supplies.
 
An alternative approach would be to have two step up converters. One 32 volts and one 40 volts. The output would be fed via a diode from the 32 volts supply and via a diode and switching P channel mosfet from the 40 volt supply. There would be almost no power dissipation in the switching device using this method.

Les.
 
One way to do this is with an audio power amp chip. If you really need to change states in less than 1 ms at 2.5 A, then you need some serious power bandwidth. Something that can make 20 W at 20 kHz will meet your risetime requirements and have enough oomph to charge/discharge whatever the load is.

ak
 
Status
Not open for further replies.

New Articles From Microcontroller Tips

Back
Top