Underrated transistor?

[earckens]

In the schematic attached there is transistor Q5 (C1815 SOT-23, spec sheet attached) that in theory should withstand the conditions to which it is subject:
Vcb max 60V, Vce max 50V, Ic max 150mA, Pc max 200mW.

When running this pcb in test modus, with a small 1000µF 63V condensor, all runs ok, hours on end.

Yet when even one single 4700µF 63V condensor is connected, the circuit output goes into oscillation after some time (ie output voltage fluctuates +/- 5 to 10V), and, as was the case just now, suddenly burns out Q1 (and takes the LM723 with it).

Condensor Vcc = 46V.

What could be going on, except for the fact that in theory things should be fine?

How to remedy?

EDIT: C7 has been replaced with a 100µF condensor, the 18Vdc line is free of any ripple.

 

[alec_t]

What could be going on

Just a guess, but if the power supply switches off (or goes into violent oscillation), Q1 base voltage may drop to 0V, in which case Q1 Vbe will be -7.5V. The maximum reverse voltage of the base-emitter junction, per the datasheet, is only 5V, so perhaps the junction breaks down.
Given the values of R12 and R15 the breakdown current would be small, so whether or not it would be fatal is debatable.

 

[earckens]

Just a guess, but if the power supply switches off (or goes into violent oscillation), Q1 base voltage may drop to 0V, in which case Q1 Vbe will be -7.5V. The maximum reverse voltage of the base-emitter junction, per the datasheet, is only 5V, so perhaps the junction breaks down.
Given the values of R12 and R15 the breakdown current would be small, so whether or not it would be fatal is debatable.

That sounds plausible, but where does the -7.5V come from? The zener diode is connected to Q3, while the burnt out transistor is Q1?

IMPORTANT EDIT: The zener diode is connected to Q1, the burnt-out transistor is Q5

And I cannot find a transistor with Vbe better than 5V?

Even a BD139 (spec attached) does have 5V Vbe only?

 

[rjenkinsgb]

That sounds plausible, but where does the -7.5V come from? The zener diode is connected to Q3, while the burnt out transistor is Q1?

If C12 is charged and the main power is shut off, the base may get pulled near 0V while the emitter is still near 7.5V

I doubt it's a problem as the base resistors are such high values and the chance of it happening is slim - but all you need to do to eliminate it as a possibility is add a signal diode across Q1 base-emitter to protect it if the base is negative, so anode to the emitter.

 

[earckens]

If C12 is charged and the main power is shut off, the base may get pulled near 0V while the emitter is still near 7.5V

I doubt it's a problem as the base resistors are such high values and the chance of it happening is slim - but all you need to do to eliminate it as a possibility is add a signal diode across Q1 base-emitter to protect it if the base is negative, so anode to the emitter.

And what do you think of the use of an SOT-23 for Q1? Just 200mW.

I experimented a few minutes ago with an TIP41C for Q1 (TO220, P 65W, Vce 100V, Hfe ~40), the circuit still works. But is that the right approach (besides adding the base diode)?
Or what about a BD139 (TO225, 12W, Hfe ~100, Vce 80V))?

..or is that the wrong approach?

 

[rjenkinsgb]

The maximum current it should ever pass is less than 25mA at start up and under 15mA steady state once the 18V is stable, from the 2k2 collector resistor.
And normally just 10.5V across it, so operating dissipation under a quarter of a watt.

I'd use something physically big enough to handle half a watt without getting beyond slightly warm, but it does not need any particularly high power ratings, just good gain.

I'd not use a surface mount device unless well rated and on a good area of copper for cooling.

 

[earckens]

Here are the scope signals (Edit: Q1 Vbe) on startup and on shutdown. Probes are x10 so all y-values must be taken x10.

On startup a single 1V (20mV x10 x 5 divisions) occurs.

On shutdown a downward drop of 2V (50mV x 10 x 4 divisions) occurs.

Edit: this is with 1x 4700µF 63V condensor.

 

[earckens]

The maximum current it should ever pass is less than 25mA at start up and under 15mA steady state once the 18V is stable, from the 2k2 collector resistor.
And normally just 10.5V across it, so operating dissipation under a quarter of a watt.

I'd use something physically big enough to handle half a watt without getting beyond slightly warm, but it does not need any particularly high power ratings, just good gain.

I'd not use a surface mount device unless well rated and on a good area of copper for cooling.

Ok, point taken.
Maybe a TO92 C1815 (400mW), or else a BD139 (TO225, 10W and Hfe ~ 25 to 250)?

I have it working now with a TIP41C, so a BD139 (with higher Hfe) would surely be better.. ?

 

[rjenkinsgb]

Remember power device ratings are when on heatsinks.
The BD139 is rated at just 1.25W in fee air.

I'd go with the C1815, using the highest gain one you have, and with a clip on heatsink. For a low gain device I'd change the two 47K resistor to 10K or less. With the two 47Ks across 18V, the divider current is only about 200uA.

A transistor with a gain of 100 controlling the 15mA collector current would take 150uA base current, totally messing up the divider ratio. I'd be looking for gain 400+ there, and probably lower resistor values.

 

[earckens]

rjenkinsgb Hi JRW, I messed up with the transistor ID's. When writing Q1 in post #7 that should be Q5. Same in post #3 (which is now corrected in red).

Q1 is C1815 SOT23.

To recap:
1. the scope signals are measured accross Vbe for Q5.
2. Q5 is the one that blew up

Questions:
1. Q5 replaced with a TO220 TIP41C and the circuit still working, ie output voltage is regulating. Is that Hfe critical (~25 for this transistor)?
2. Q5 alternative: BD139? Same question as in line here above: is Hfe for this transistor good enough (~50)?
3. In my post #3 I made the same mistake for transistor ID's (edited now, in red). Question: the Vbe ~7.5V negative will be on Q1, so how can this affect a burnt-out Q5?
4. In view of the above, is your statement about replacing Q1 (C1815 SOT23) with a higher powered C1815 still valid?

 

[rjenkinsgb]

OK, Q5 should not pass any high current under normal conditions, but could have when the 18V supply was oscillating?
With the configuration it's in, the hFE is not critical.

If Q1 is staying near enough cold, leave it as its - though personally I'd use something with a bit higher power rating, based on the circuit.

 

[earckens]

OK, Q5 should not pass any high current under normal conditions, but could have when the 18V supply was oscillating?
With the configuration it's in, the hFE is not critical.

If Q1 is staying near enough cold, leave it as its - though personally I'd use something with a bit higher power rating, based on the circuit.

The 18V_DC is, imho, stable: almost regardless of what appears on the Vcc line (46Vdc) its voltage will not change.

Although, I likely will change Q1 from an SOT23 to a TO92 (400mW instead of 200mW).

Q5: I rather think of the 46Vdc to present transient spikes.
As well as the two scope visuals I posted above affecting Q5: maybe not that critically high voltage levels, Q5 being an SOT23 (and it having been burnt a few days ago) does not make it a robust package.

So, if Hfe for Q5 not being critical, maybe time to move to a bigger transistor package? Any other ideas?

 

[AnalogKid]

In post #1 you mention a 1000 uF cap. The schematic is very difficult to decipher, but there is none that I can find on it. Also, no reference designator, so I have no idea where this capacitor is connected.

If you are talking about hanging a large value capacitor on the output, then this probably is a simple gain margin / phase margin issue. Most opamps cannot handle large capacitive loads, and this is a particular problem for circuits with a large power bandwidth, such as audio power amplifiers and voltage regulators.

Separate from that, the circuit seems very complex for the task. What are the requirements for the circuit? Min / max output voltage, min / max output current, current limiting type, etc - ?

ak

 

[earckens]

In post #1 you mention a 1000 uF cap. The schematic is very difficult to decipher, but there is none that I can find on it. Also, no reference designator, so I have no idea where this capacitor is connected.

If you are talking about hanging a large value capacitor on the output, then this probably is a simple gain margin / phase margin issue. Most opamps cannot handle large capacitive loads, and this is a particular problem for circuits with a large power bandwidth, such as audio power amplifiers and voltage regulators.

Separate from that, the circuit seems very complex for the task. What are the requirements for the circuit? Min / max output voltage, min / max output current, current limiting type, etc - ?

ak

I am so sorry for this confusion, please accept my apologies.

The 4x 4700µF caps are not used in my prototype testing. Instead I did use a 1000µF 63V cap.

The 1000µF cap is placed where the 4700µF caps are shown in the schematic.

There is no capacitive load on the output of this circuit.

Circuit requirements:
1. minimum output voltage: 0,00V
2. max output voltage: more or less 40V
3. max output current: 3A
4. current limiting: using LM723 regulation.