Following rumors found on audio DIY forums, we measured a number of power types. Some weren't very bad, but they varied greatly from part to part. Two power parts made it onto our table, in good spots: BD437 TO-126 and BU406 TO-220, but awwkk, they have horrible measured beta. So instead I suggest a Sziklai power stage.
OK thanks Win. I mostly just pick transistors from what's in stock.. and not from any deep understanding of knowing what the heck I'm doing. :^)
George H.
0V, actually. Ground the collector.
Does anyone add some surface conductors, like a solar cell?
Couldn't you do a current/voltage graph of the b-e junction and figure out Rbb from that? Would the ohmic part of the curve be Rbb?
OK I'm going to order some of the ZXT851/ 951 (not a lot of stock at DK) and the ZXT450/550, I'll report back with results.
Thanks everyone. George H.
It's possible to put an indium-tin oxide layer on top, but it costs extra and the stack is much harder to AR coat.
I'd happily settle for a higher-doped epi. We're investigating the cheapest way to add immersion lenses to plastic-package photodiodes.
Cheers
Phil Hobbs
Friday closure. I measured beta of a few of the transistors (1,10,100 mA) (using JL's circuit above, collector grounded) and found no correlation between beta and the noise. Only a few data points, but the noisest 2n4401 had a higher beta, than a quite one...
George H.
Looks like you have to make a noise selection meter. If you are after noise in the audio range maybe use a low noise amp, a PC and the Daqarta software.
Beta goes up with temperature, so that's not too surprising.
Cheers
Phil Hobbs
Probably base resistance, Rbb. It would be interesting to check b-e with a DVM on the ohms or diode range, and see how that correlates to noise.
Ahh, OK. I've got these wired in to my white proto board. (I looked at the emitter voltage with a scope and saw no bad oscillations.) You want me to check V_be. I can do that, but you have to wait till Tuesday! (My kids have voted for shiskaba.. (I don't know how to spell that word.) Chicken, beef, onions, peppers, tomatoes, stuck on a stick and marinated in Chiavetta's (I was going to say you can't get any... but amazon.),
What's the best beef for shiskaba?
George H.
That sounds plausible, because it's oversimplified. Lower ib makes higher percent shot noise in ib, but not necessarily in the Ic, because Ib absolute shot noise is lower even if percantage is higher.
Extra words in the description really help in cases like this...
Oughtn't the voltmeter be across the feedback resistor? That voltage is reciprocal to beta.
Win, could you recommend transistors for Sziklai power stage for ~450mA-500mA cap mult? Thanks!
For good parts, with high beta and low r_bb', the effect is too small to give good answers.
I'd use a high-beta, low r'bb part like an NPN ZTX618 or FMMT618. Run it at 20mA with a 33-ohm PNP base-emitter resistor, and a PNP power transistor of your choice.
AFAIK it is spelled Shish Kabab. Best to ask someone from Turkey though. I think shish translates into skewer. My mountain biking buddy made us really nice sturdy skewers for that from stainless steel.
Tenderloin or, if you want to really splurge, filet mignon. I prepare them as usual over wood fire.
had an output noise of ~>1.5 nV/rtHz.
) )
in
for
rce
What exactly is the circuit? Is the transistor an emitter follower regulato r for switched capacitor step up? If that's the case then a higher beta mea ns more feedback loop gain and therefore more attenuation of the transistor contribution to output noise. That is probably the correlation you're seei ng. As for temperature, transistor beta fades at 1%/oC, which explains the higher noise with temperature, in addition to the kTRB dependence.
Thank you, Win! Did you mean something like this?
Have something to add? Share your thoughts — no account required.
Ask the community — no account required