Emulating Open-Collector operation with TTL 74LS138...

torsdag den 6. oktober 2022 kl. 08.24.33 UTC+2 skrev John Robertson:

PNP ..

mosfet, gate to 5V, source to 138, drain is your output

So, a cascode-like second transistor atop the OC output. Good idea.

Can also use bipolar NPN, with the base through a resistor to +5V; has slightly lower pull-down.

Switch it to a 139 and use a 2N7002?

Cheers

Phil Hobbs

If your Vbb is always within 30V, you can replace the 74LS138 with a 7445, although not the same pinout. If the load current is no more than 80 mA you could also dump the TIP125. But since the TIP125 is rated for amps, I expect 80 mA is not enough. The 7445 is not easy to find, however Jamesco claims to have them. I assume DIP is ok?

Using a BJT in a "collapsible cascode" like that is liable to produce a nasty ~1 us spike of nearly the full supply voltage on the OC output, due to charge storage in the cascode transistor.

And the capacitances of the MOSFET can do the same, except not as badly.

Either way, a Schottky diode from the '138 output to +5 would be prudent to avoid stressing the ESD structure on the chip.

Cheers

Phil Hobbs

A cap won't do what you want. It will charge up and the 138 won't be able to drive the TIP125. Unless... you add biasing so the 138 output was at say, 10V while open. The cap would charge up to 10V while idle. When the 138 pulls low, the cap will provide the current to drive the TIP125 for 10 ms. I think this would need about a 10 uF cap which is available in ceramic. It depends on the base drive current of the TIP125, but that should be pretty minimal because of the high gain.

This will isolate the voltage, but with no current gain. Since the TIP125 will drive amps and is in a TO220 case, I'm thinking the current gain is required.

7445 was in the original design, the board I'm dealing with was redesigned and installed a 138 instead of a 45 - I am trying to save the boards by modifying the drive to the PNP TIP125s.

I am considering simply making a small 74138 to 7445 PCB - might be the easiest.

Next board will have 7445s. I have reasonable source of 7445, the number needed is under 100 for the entire run of the boards AFAIK. Digikey has around 1500 @ $4 each, and there are other sources for less.

Thanks,

John :-#)#

Sorry, for some reason, I thought the '138 was an open collector part. I don't follow what the original circuit was supposed to be. I'm thinking you are designing a circuit from scratch??? The '138 could never have been used to drive the TIP125 in this arrangement.

To drive a high current load from the 20V supply, you either need a higher voltage output than the '138 offers, or you need to use transistor as a low side switch, or you need two transistors to provide the voltage shifting.

The 7445 will work if you are happy with using such an outdated part.

Can't see that working...this is switching around 20V.

Here is the circuit:

What I want to do is add a single component in series with RC0/RC3/RC6's

1KR to enable the 138 to drive the PNP TIP125 and isolate the output of the 138 when output is High. MOSFET could work, but switching surges might be excessive. Leaning towards making small PCB to convert 74LS138 to original 7445 as the simplest solution. Space is a problem around the transistors you see, no room for additional drive transistors as this is finished PCB.

Thanks,

John :-#)#

Repairing a new PCB design...and yes, the circuit originally used a

7445. Kinda stuck for this PCB run - next revision will use the 7445 (or MOSFETs) in place of the 138s...

Thanks,

John :-#)#

Well, this is to fix 5 x PCBs we are stuck with at the moment, next run will fix the drives - either 7445s or 138 with P-Channel MOSFETs and drivers.

Circuit I am trying to fix with fewest components - one x two legged device preferred per 138 output:

VLAMP is roughly 20VDC.

Thanks,

John :-#)#

That's not clear. Got a schematic?

Any idea how the '138 board was supposed to work? Was it intended to drive something different than the TIP125?

You might be able to make the capacitor drive work. It would need a few 10s of uF which you can get in a ceramic cap. Since the '138 drives a 5V range and the TIP125 requires a 2.8V swing at R2, you should be able to make it work. It will probably require pullup resistors on the '138 outputs to restore the proper voltage on the cap when not driving, maybe 1k. The circuit will power up driving the TIP125 on until the cap is charged up. Is that a problem?

This would be sensitive to voltage transients on the Vlamp rail. Fast voltage transients will trip the TIP125 on, but if they are just powering lamps, it likely won't be visible.

it is?

and the three missing legs are where?

TIP120, TIP121, TIP122 (NPN); is the right drawing

TIP125, TIP126, TIP127 (PNP); is the left drawing

It would seem a combination of a high side PNP driver and a low side NPN level shifter would be a useful, single package device. Is there a reason why this is hard to make on a single die?

Yes:

Think going to a 7445 sub-board will be best way around this. It was the designer who made an error on using the 138 and I am simply trying to rescue 5 populated boards...blame the Scottish side of my family!

John :-#)#

The MOSFET cascode thing is pretty good actually, if you put a diode to the supply or want to live dangerously and rely on the ESD protection to discharge the gate. If it's a matter of using the boards or tossing them, there's not much to lose by using barefoot FETs.

2N7000s are still around IIRC, and they're TO-92. There are multiple-section devices such as the

Cheers

Phil Hobbs

'138 is push-pull, though very wimpy high side