Collector-Emitter breakdown voltage of 555 discharge Q

[Connoiseur]

Hi everyone!

I am building a PWM controller for a small heating element (open-loop). I was thinking of using the following circuit (R6 and R8 will be replaced by pot). Unfortunately I couldn't find any data about the Collector-Emitter breakdown voltage of 555 discharge transistor in any manufacturer's datasheet. They just mention the max input for vcc, trig,thres and cont. Has anyone an idea?

Thanks

[Z80]

Hi, I'm not a fan of that arrangement, the discharge transistor in the 555 is there to discharge the timing cap.  As you are just switching a heater, it would be easier to use a low side mosfet.  The attached circuit will work fine and if you are using a small mosfet you could omit the two driver transistors.

[Zero999]

Hi everyone!

I am building a PWM controller for a small heating element (open-loop). I was thinking of using the following circuit (R6 and R8 will be replaced by pot). Unfortunately I couldn't find any data about the Collector-Emitter breakdown voltage of 555 discharge transistor in any manufacturer's datasheet. They just mention the max input for vcc, trig,thres and cont. Has anyone an idea?

Thanks

You're also most likely exceeding the maximum gate voltage rating of the MOSFET, which is usually less than 20V, without looking at the data sheet for the transistor you're using.

There may be a reason you want high side, rather than low side switching. You could add an NPN transistor in common base and an 10k emitter resistor which will reduce the gate voltage to a safe level.



I haven't done any calculations/simulation. This circuit will be quite slow. Check that the MOSFET is switching fast enough to avoid exceeding its maximum power rating, due to the switching losses.

[capt bullshot]

You should expect the Collector-Emitter breakdown voltage beeing equal the abs. max rating for the 555 power supply pin.

[David Hess]

I am a little surprised that the datasheets do not include the Vceo of the open collector discharge transistor.  You could just measure it by monitoring the collector leakage as the voltage at the collector is raised while the transistor is off and normal Vcc is applied to the 555; use a large value series resistor to avoid damage.

[floobydust]

I think the 555 IC's process is a 18V max, so I would not expect the discharge transistor to do above that.

Your PWM circuit has other problems, I would not apply 24V to gate of a mosfet as 20V is generally the max. rating. I'd add a 18V zener and use a separate transistor to switch the gate.

[David Hess]

I think the 555 IC's process is a 18V max, so I would not expect the discharge transistor to do above that.

I wonder if it was built on a logic process.  Offhand I do not remember a 18 volt analog process from that time but the various detailed schematics including the Signetics schematic show PNP current mirrors and PNP differential pairs and I would not expect PNPs on an old logic process.  (1) Further, all of the 18 volt Signetics designs from 1972 that I checked include NPN transistors only.  Would they have  built the 555 on its own process?  If PNP transistors were available, I would have expected to see them in something else.

(1) The last TTL processes like FAST included PNPs but were considerably later than the 555.

[floobydust]

It must be an analog process, and I'm not sure what limits the IC's voltage- the Early effect, dielectric breakdown, bias changes etc.

Hans Camenzind has a book Designing Analog Chips he mentions to get HV you need more room for the depletion regions {pdf pg190}. So I think the transistor size is the limiting factor. He also mentions you can add process steps {pdf pg27}.

From 555 die pictures, the discharge transistor appears the same (large) as the output transistors. No room for bigger.

[ali6x944]

As a conservative guess I would say the transistor is discharging a cap at 2/3 VCC, so I would say for an NE555 discharge transistor can handle between 12v to 20v max...
But as many have suggested I don't think it's a good idea to use it in this way...

[T3sl4co1l]

All the transistors will be the same voltage rating, yes.

(Which, by the way, means the PNPs are also rated for 18V emitter-base -- because they're lateral PNP in that design, so are actually NPN collectors, touching close enough to get an awful, but usable, hFE. )

The breakdown mechanism will be avalanche.  Most chips of the era were modestly doped, giving a 30-40V limit.  A heavier doping would give lower rating, higher leakage, higher current density and shorter carrier lifetime (faster), which is probably what was used for the 555.  Gold doping would give very low rating, and more leakage and speed.  (AFAIK, 74 series TTL was your basic 30V NPN process, with gold doping bringing it down to about 8V breakdown.  2N2369 might seem kind of odd, as it's a gold-doped switching transistor rated for 15V -- but if you suppose it would've been a 60-100V bipolar without the gold doping, this seems quite reasonable.)

Taking a quick look at several datasheets, I'm surprised I don't see any individual pin voltage ratings; a few specify input voltage limits (0 to VCC), but leave OUT and DISC undefined.  Perhaps this was typical of the era, when equivalent internal diagrams were common, and processes were cheap and simple (all the same kind of transistor).  Modern chips do not typically show internal diagrams, ESD diodes, and can have many dependent voltage limits (like a bootstrap gate driver's 18V supply limit on top of a whopping 600V high side limit).

Tim

[Connoiseur]

Thanks you guys for your valuable inputs.

It does seem like a bad idea to build this circuit, will go with a low side switch.

Hi, I'm not a fan of that arrangement, the discharge transistor in the 555 is there to discharge the timing cap.  As you are just switching a heater, it would be easier to use a low side mosfet.  The attached circuit will work fine and if you are using a small mosfet you could omit the two driver transistors.

This was going to be a quick and dirty hack-together project with whatever I could find in first sight, did not want to use the IRFP250 which I had set aside for some other project, hence the PMOS.

All the transistors will be the same voltage rating, yes.

(Which, by the way, means the PNPs are also rated for 18V emitter-base -- because they're lateral PNP in that design, so are actually NPN collectors, touching close enough to get an awful, but usable, hFE. )

The breakdown mechanism will be avalanche.  Most chips of the era were modestly doped, giving a 30-40V limit.  A heavier doping would give lower rating, higher leakage, higher current density and shorter carrier lifetime (faster), which is probably what was used for the 555.  Gold doping would give very low rating, and more leakage and speed.  (AFAIK, 74 series TTL was your basic 30V NPN process, with gold doping bringing it down to about 8V breakdown.  2N2369 might seem kind of odd, as it's a gold-doped switching transistor rated for 15V -- but if you suppose it would've been a 60-100V bipolar without the gold doping, this seems quite reasonable.)

Taking a quick look at several datasheets, I'm surprised I don't see any individual pin voltage ratings; a few specify input voltage limits (0 to VCC), but leave OUT and DISC undefined.  Perhaps this was typical of the era, when equivalent internal diagrams were common, and processes were cheap and simple (all the same kind of transistor).  Modern chips do not typically show internal diagrams, ESD diodes, and can have many dependent voltage limits (like a bootstrap gate driver's 18V supply limit on top of a whopping 600V high side limit).

Tim

It might have appeared to the manufacturers that no one's ever going to put it in a circuit with different voltages on these pins OR they expected too much know-how about chips from people who use the 555. 

[David Hess]

It must be an analog process, and I'm not sure what limits the IC's voltage- the Early effect, dielectric breakdown, bias changes etc.

The early effect and bias changes would not apply to the open collector output transistor used for the discharge pin.

The Signetics 567 with a Vcc of 9 volts and Vce of 15 volts might have been on the same process as the Signetics 555.

5, 15, and 30 volt Vce are common on TTL open collector output drivers.  Maybe the 555 and 567 were built on the same process as 15 volt TTL although I only see 5 and 30 volt Signetics TTL parts.

Like I said, just test it.  Measure the voltage to produce say 1uA of leakage with a 33k series resistor.  I was never a fan of the 555 so I do not have any handy to play with.

[Zero999]

I agree with all of the above but one important thing to note is that we're talking about the bipolar 555 timer here.

CMOS 555 timers such as the ICM7555 and TS555 are very likely to have an ESD protection diode connected between the drain of the discharge transistor and VDD.