TL783 - are there any newer high-voltage linear regulators?

There are a few 60V parts around. And there are tricks to help out a lower voltage reg.

Or make your own around a depletion fet. What are your numbers?

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

On the plus side, if you can handle the dropout voltage, it's a pretty good chip.

Cheers

Phil Hobbs

-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203 Briarcliff Manor NY 10510

hobbs at electrooptical dot net

???

Tim

-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website:

A rare name.... I worked with a guy called Nagle in Goppingen ( Germany) in 1975..... any relation to you ????

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus

*Shrug*, I know a Daigle. The Simpsons have a Lindsey Naegle character. Can't be too rare.

(but I'm not one to judge, there are hundreds of Tim Williamses) :)

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: http://seventransistorlabs.com

It's the huge dropout voltage that's the problem. There has to be about 15V between IN and OUT for it to work.

Linear Technology has some "high voltage" parts, but they top out at 80V.

On Semi has a NCP781 150 V, 100mA Very High Voltage Linear Regulator, but the output voltage cannot exceed 15V.

Some other vendors have 60V parts. The TL783 is the only part I can find that can regulate up to 120V.

There are some parts good up to 450V, but they require big input/output differentials. They're for getting a little bit of

+5 directly from the AC line.

(The application is a driver for Teletype selector solenoids. The upstream circuitry charges up a 2uf cap to 120V to provide pull-in against the huge inductance, then dumps that into the output. After discharge, a sustain supply takes over. I have all that working for 220 ohm 5.2 henry selector magnets.

The problem is that there are several different magnets, with different resistances and inductances. All of them need

60mA, and the big jolt from the cap will overcurrent the low resistance solenoids. The higher resistance ones don't need this protection. So I want to put current limiting to 60mA on the output. It has to work over the voltage range 12-120VDC.)

John Nagle

Well pffbt, put a DMOS on there and be done with it!

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: http://seventransistorlabs.com

Gotta be careful with the protection circuitry though, especially driving inductors. Ask George. ;)

Cheers

Phil Hobbs

Thanks. Looking at

for starters.

John Nagle

But, a voltage regulator has LOW output impedance, and to drive an inductor fast, you want HIGH output impedance. Like, return the negative side of the solenoid through an NPN collector, with base grounded and emitter through 200 ohm limit resistor to -12V.

tive

You need high voltage to get a big dI/dt. Low impedance doesn't hurt and ma y help.

Cheers

Phil Hobbs

How important is cost? LTC makes some nice HV opamps. LTC6090 140V rail to rail output. ~$4 for 100.

George H.

How much power are you talking about? It's a simple matter to make your own with discretes, and your 125V does not suggest that state-of-the-art efficiency is a requirement.

puts too much resistance in the circuit. To get 2V at the gate requires a sense resistor of about 33 ohms. The load is only 55 ohms. I'm going to waste a third of the energy in the sense resistor.

Is there some part that contains one of those DMOS FETs and a gate amp to make it a regulator?

John Nagle

gative

may help.

A voltage regulator (series pass type) just steals some of the high voltage , which (unless something's odd in the coil's insulation resistance) is a needless addition to the circ uit. Current regulation is more appropriate than voltage regulation.

To really drive a solenoid fast, a second inductor and current source to bi as it is a good power supply, better than a filter capacitor and voltage source. Th e constant power drain from the current source, though, is an energy hog.

Well, hopefully not just a third, but up to a factor of a hundred: you gave a 10:1 spread in voltage, and a fixed current limit spec.

Is that not what you were expecting?

Or to put it another way, if you can give a _complete_ description of the required operation, you'll get more sensible ideas. :-)

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: http://seventransistorlabs.com

The idea is to use a PMOS wrapped around the regulator--drain to +HV, gate to REG_INPUT, source to REG_OUTPUT, with a resistor from drain to gate to supply bias. You pick a MOSFET with a high enough threshold that the regulator's dropout is not a problem. It's been done with PNPs wrapped round IC regulators from the very early days of ICs.

You'll need to protect the regulator from various fault scenarios.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

Belay that--I have way too much going on. In a low voltage version, it's source to +V, gate to REG_INPUT, drain to REG_OUTPUT, resistor from source to gate. The IC regulator only passes enough current to bias the FET by way of the resistor.

In the HV version it's probably simpler to use an op amp.

(I need a vacation really really badly ATM.)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

Complete description:

Full design in KiCAD, including the board layout and SPICE model.

This works fine for a 220 ohm 5 Henry solenoid as the load, with about 60mA being maintained through the cycle. The circuit charges up a 2uf cap to 120V, and dumps that into the solenoid. Then a 12V sustain supply takes over and holds the solenoid in. This happens on each bit time, which is at least 22ms.

All this is powered from a USB port, which is why it's a bit complicated. It draws about 300mA at 5V.

Now that I have all that working, I just want to throttle it back for a different selector magnet which is 55 ohms with a smaller inductance, but still wants a 60mA signal. So I'd like to put a 60mA current limiter on the output. There are a few different selector magnets on various old Teletypes, all designed for 60mA constant current. The output will start at 120VDC and drop to as low as 4V as the inductor charges. But the current is supposed to be constant.

(The traditional Teletype power supply is an isolated 120VDC supply with a 2K 10W resistor in series, with over 90% of the energy going into heating up the ballast resistor.)

John Nagle