ZTX 109 Replacement

hello,

It's been a while since I've posted. I'm helping a friend repair the mute automation on his Allen and Heath Saber. The battery leaked and damaged some components and parts of the PCB.

One of the fallen  is a pair of ZTX 109 BJT NPN transistors. I think these were made by Diodes Incorporated. It appears these have been out of production for a while, so I am looking for a replacement. It looks like the ZTX 449 is a suitable replacement, but I was hoping someone could compare and offer a second opinion.

A couple links to the transistors in question:

http://www.farnell.com/datasheets/49571.pdf

http://www.mouser.com/ds/2/115/ZTX449-92266.pdf

Thanks.

The BC109 was IIRC a jelly bean general purpose (does everything) part...

The 449 looks like it covers most of the specs.

In situations like this it is perhaps worthwhile to look at the circuit where the parts are used to see  what parameters may be important.  Or you could just try a few and see if they work.

Back when I was active designing circuitry I had a handful of common part numbers I would use over and over for general purpose applications...  2n3904/3906.... 2n4401/4403, MPSA05/A06, etc...

Unless it is exposed to unusually high voltage, or high gain requiring low noise it is probably OK to use the generic device du jour.

JR

hi John,

First, post the schematic ... great advise.

It's a little fuzzy but I hope you can make it out. If not, I can redraw it. You can see the pair of ZTX 109s. The collector of the second connects to A15 to IC8 and on to IC3.

These components were soldered on the underside of the PCB so I'll need to find something in a TO-92 package. Not the can style.

Thanks again.

OK that schematic looks incomplete or incorrect (not uncommon with old hand drawn schematics).  That is part of the fun..

The open collector of the second 109 looks like it is generating an inverted version of A15

If the schematic is correct,,, the 15V rail should turn on the first 109 , which will turn off the second one.  Perhaps a power off mute, or power on reset/clamp... that holds the second collector low until the power supply rises to full 15V.

If this is what the circuit is actually doing (just speculation on my part), the Vbe of the first 109 will affect the turn on (off) voltage threshold.

The data sheet for the 449 specs a scary 1V Vbe max, but looking at the Vbe curve later in the data sheet looks more typical.

I will speculate that it will work, but that ASSumes the circuit is doing what I guessed (turn on/off clamp).  Worst case you might have to tweak the voltage divider feeding 1st 109 if base sensitivity is too different. 

If you only have one good 109 put it in the 1st position, the second 109 is less critical. Also maybe look for extra parts not shown in the schematic (like another resistor).  8)

JR

John,

I doubled checked the actual PCB and the schematic is correct. I made a quick sim of the circuit using BC547B transistors and the results are  attached. I did a sweep from 0 to 15V simulating a power up.

The blue line is the voltage at the base of Q1 and the green line is the voltage at the base of Q2.  Am I reading this correctly that Q2 is "on" during power up then "off" once power up is complete (14V according to the sim).  It's a little cut off on my previous attachment, but the collector of Q2 goes to IC7 pin 19.

IC7 is a SN74LS245N, an octal bus transceiver. The description of pin 19 says "Active low output enable; Low = all channels active, High = all channels disabled" so if I understand correctly, the circuit in question is going "high" to disable all the channels during power up and "low" to enable them when the power up threshold is met (14V according to the sim).

Any thoughts? Also I think the BC547 or BC549 will do the trick. They have the proper Vbe of 0.7V.

Greg said:

John,

I doubled checked the actual PCB and the schematic is correct. I made a quick sim of the circuit using BC547B transistors and the results are  attached. I did a sweep from 0 to 15V simulating a power up.

The blue line is the voltage at the base of Q1 and the green line is the voltage at the base of Q2.  Am I reading this correctly that Q2 is "on" during power up then "off" once power up is complete (14V according to the sim).  It's a little cut off on my previous attachment, but the collector of Q2 goes to IC7 pin 19.

IC7 is a SN74LS245N, an octal bus transceiver. The description of pin 19 says "Active low output enable; Low = all channels active, High = all channels disabled" so if I understand correctly, the circuit in question is going "high" to disable all the channels during power up and "low" to enable them when the power up threshold is met (14V according to the sim).

Any thoughts? Also I think the BC547 or BC549 will do the trick. They have the proper Vbe of 0.7V.

Click to expand...

try them...

Transistor Vbe is not a firm voltage threshold, but a squishy just about or more or less voltage around 0.5-0.7V  The actual voltage will change with temperature, current, and if you look at it crosseyed.

If the sim says it works it must be OK....  ;D  (I haven't used a sim yet and I'm too old for new tricks).

JR

Greg said:

Am I reading this correctly that Q2 is "on" during power up then "off" once power up is complete (14V according to the sim).  .

Click to expand...

This circuit is so basic it would work with almost any small-signal NPN Si transistor. You may run several sims with different transistors and see they all do just about the same. The only condition is that Hfe (beta) is high enough.

Also I think the BC547 or BC549 will do the trick. They have the proper Vbe of 0.7V.

Click to expand...

ALL Si transistors have a Vbe of 0.7V, from the smallest  to the big power transistors. As JR mentioned, 0.7V is a shortcut for a logarithmic curve that seems to flatten at about that value. Since this voltage is dependant only on the nature of the semiconductor material, I wonder why anybody cares to include that in datasheets...

Thanks for the info. It's a rather basic exercise but fun nonetheless. It's been a while since I've fooled with BJTs (probably since electronics class as an undergrad).

I'll get some parts on order. The PSU on this board was damaged and has to be totally rebuilt as well.

Thanks again!

abbey road d enfer said:

ALL Si transistors have a Vbe of 0.7V, from the smallest  to the big power transistors. As JR mentioned, 0.7V is a shortcut for a logarithmic curve that seems to flatten at about that value. Since this voltage is dependant only on the nature of the semiconductor material, I wonder why anybody cares to include that in datasheets...

Click to expand...

You're correct that Vbe is a function of the semiconductor material, but in a finished device, the extra base resistance can skew the curves slightly. For some people, or for extra small or lightly doped devices, it's helpful to know a more accurate V/I curve for a particular device's base-emitter junction. What would be more clever would be to specify Rbb, but for those who want something simpler, a datasheet graph will be more popular.

Monte McGuire said:

You're correct that Vbe is a function of the semiconductor material, but in a finished device, the extra base resistance can skew the curves slightly. For some people, or for extra small or lightly doped devices, it's helpful to know a more accurate V/I curve for a particular device's base-emitter junction. What would be more clever would be to specify Rbb, but for those who want something simpler, a datasheet graph will be more popular.

Click to expand...

The data sheets that Greg posted included the Vbe curves...  The fact that they are curves reinforces that they are not a single static voltage.  The generally don't publish Rbb for general purpose devices, but do publish it for low noise devices where it can contribute to noise voltage.

I did a lot of work using Vbe for log conversions. In a very old design (Loftech TS-1), the Rbb of the small geometry devices in the transistor array I used was enough to introduce a linear error to the log Vbe.  I estimate my TS-1 read a fraction of a dB high at +20dBu because of the base current times Rbb.  8)

JR

I just wanted to provide an update. I rebuilt the power supply section and the circuit in question with BC547s.

I pulled the datasheets on IC8 and IC3. I incorrectly stated the function of these ICs. IC8 is a 74LS05 "Hex Inverter with Open Collector Outputs" and IC3 is a MK6264LN-120 "8-Bit CMOS Static RAM." It's worth noting the RAM has a power down standy mode whenever E1 Bar goes high. E1 Bar is Pin 20 on IC3 and Vbb is 4.3V (5V - 0.7V drop across D2) when the supply is on an 3.6V from the battery when the supply is off.

On my bench with the 15V supply on (board is on), the collector of the second BC547 measures 4.5V. This signal in inverted by IC3 and  E1 Bar is low. When the I turned off the bench supply, E1 bar measures 3.6V (from the battery) and E1 bar is high essentially putting the memory into standby.

I presume when the battery and psu failed on this board, the RAM chip wasn't working and that's why the mute automation went nuts. It was behaving as if it had no memory and when I'd push a mute button, other channels would randomly mute.

I'm going to install this board back into the console this week and see if we can get it back up and running.