The usual circuit (found widely across the internet) for a true random number generator's (TRNG) entropy source is a reverse biased transistor like so:

Reverse biased transistor entropy source

Q1 is typically a 2N3904 as it has been found to be quite noisy the wrong way round running in breakdown mode. +ve is typically in the order of 11V. This topology works well initially.

I came across someone who had built a TRNG and experienced problems whereby "after about 3 months of use the noise signal would drift." I've seen other references to long term degradation of such a topology but can't seem to locate them at present.

I have the sense that continuously shoving electrons the wrong way up a transistor might be detrimental, but have not found any references to prove this one way or another. After all, this is not the principal use case for a transistor.

Could the junction suffer gradual and irreversible damage thereby altering it's noise characteristics? Zener diodes are another situation altogether as breaking down is their modus operandi.

Three references on E.SE discuss reverse biasing, but not long term stability:

Understanding reverse biased PN junction

Does a reverse-biased P-N junction create quantum noise?

Is my avalanche noise “random?”

Is a reverse biased transistor stable in the long term?

  • \$\begingroup\$ That 'reverse biased transistor' is an avalanche diode. Its transistor properties are irrelevant in this connection. So, the manufacturer won't stand behind its aging for this purpose. \$\endgroup\$
    – Whit3rd
    Feb 27, 2017 at 4:54
  • \$\begingroup\$ @Whit3rd Could you please clarify the last sentence of your comment vis a vis "stand behind" & "this purpose"? \$\endgroup\$
    – Paul Uszak
    Feb 27, 2017 at 14:12
  • \$\begingroup\$ If not specified otherwise , it is not guaranteed. The maximum current for specifying Vr minimum is ONLY @10uA with a Vr(min)=-6V so passing items may be -10V but if they draw more than 10uA this exceeds their spec for Vr so any application that uses this is not supported and known to fail such as at 100uA so they can not support any application 10~50uA with any certainty or "stand behind it" The same is true for LED's Red is rated at 1uA, White @ 10uA and all LED's are rated at -5V absolute MAX. They may also be good for noise current but risk junction failure. \$\endgroup\$ Feb 27, 2017 at 17:37
  • \$\begingroup\$ I have observed junction capacitance rise from early failures on LED's exposed to >-5V by sweeping AC with large series R and using XY scope for VI comparisons with damaged and good LEDs. The rise in capacitance occurs with a bridged nanogap in the insulator between conductors when reverse bias caused by a high flux density (heat) in a smaller zone, which causes structural failure of the semiconductor. THis Avalanche effect is quite different from Zener diodes designed to withstand large reverse currents with noise. \$\endgroup\$ Feb 27, 2017 at 17:43
  • \$\begingroup\$ @PaulUszak: the manufacturer of the transistor has taken precautions against aging, assuming normal biasing of the transistor. When you reverse the base, and leave it like that for long times, the electric fields at the surfaces are such that contamination may move around. The lifetime of the device may depend on details of surface passivation that are not tested with the base in avalanche. So, an ONsemi part and a TI part and a Zetex part may be entirely different. I've seen base avalanche cause rapid parameter changes. \$\endgroup\$
    – Whit3rd
    Feb 27, 2017 at 21:34

1 Answer 1


No they are not stable.

  • but stability is improved at lower currents <=10uA from my research. *

The current gain Ic/Ib of bipolar transistors strongly decreases when the oxide over the emitter-base junction is damaged. The results obtained with different stress conditions lead to point out two degradations processes: the main process consists in the increase of the recombination in the vicinity of the space-charge region in the emitter-base junction.



The above report was based on the 2N2222 which has a lower reverse breakdown threshold than the 2N3904 and also more noise voltage, but perhaps faster degradation. I have yet to find a report on the 2N3904 so an uncontrolled avalanche current is unstable with a small resistor

Transistor I=1μA 10μA  100μA   Noise(I=10μA) For I=[5-50]μA noise:
BC107B  9.30V   9.30V   9.28V   200mVpp     inc., max=10μA, dec.
BC548A  8.44V   8.46V   8.45V   100mVpp     constant
BC547B  8.23V   8.22V   8.23V   100mVpp     decreases
BC547C  8.36V   8.35V   8.34V   120mVpp     decreases
BC546B  8.19V   8.21V   8.19V   120mVpp     inc., max=10μA, dec.
2N3904  10.82V  10.80V  10.76V  400mVpp     constant
2N2222A 7.20V   7.25V   7.23V   440mVpp     constant

Ref http://holdenc.altervista.org/avalanche/index.html

enter image description here The above uses a constant current source unlike Paul Campbell's rng2.0 which relies on avalanche current in base * hFE limited by Collector R = 3.3k to 20V so if hFE is only 100, diode can be (20V-1Vsat)/3.3k/100(hfe)= 58uA which would be excessive.

enter image description here

Given the initial question design saturates the noise into a digital signal and noise can vary in current with temperature as well as spectral bandwidth it is not stable and the threshold above 10uA where failure rapidly degrades is unknown, limiting the reverse current to the Mfg's test criteria for Vr would make most sense.

The next question is how stable does it need to be? What is life span criteria and repeatability to NIST scores? What are the test results over ambient range?

Increasing the coupling capacitor size may increase the max number of consecutive 1's, a figure of merit by extending the number of decades of frequency range at the expense of startup delay or sensitivity to supply regulation drift.

  • \$\begingroup\$ Does there exist a TRNG that's stable then? \$\endgroup\$
    – Bradman175
    Feb 27, 2017 at 1:04
  • \$\begingroup\$ @Bradman175 Let's be very clear. The circuit fragment in this question is not a TRNG. It is an entropy source, which means that it is a source of unpredictable electronic signals. That's all. To produce uniformly distributed random numbers takes more circuitry in conjunction with software based randomness extraction. I'm not aware of any commercial TRNG that operates entirely in hardware without software extraction. \$\endgroup\$
    – Paul Uszak
    Feb 27, 2017 at 1:14
  • \$\begingroup\$ White noise can be created with Zener diode current, Avalanche diodes and IMPATT Diodes \$\endgroup\$ Feb 27, 2017 at 1:25
  • \$\begingroup\$ But considering Noise is now the Signal now, and S/N ratio will determine your entropy by equal power distribution. Any spurious resonances or recurring interference noise will degrade the entropy. This can best be determined with a spectrum analyzer. with >60dB SNR \$\endgroup\$ Feb 27, 2017 at 1:33
  • \$\begingroup\$ @TonyStewart.EEsince'75 You appreciate the magnitude of your answer? This would mean (as I suspected) that virtually every TRNG design on the internet based on transistors is unstable in the long term. It also trashes the reliability of such commercial devices as OneRNG, entropykey, TrueRNG, ChaosKey et al. \$\endgroup\$
    – Paul Uszak
    Feb 27, 2017 at 1:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.