Thank you all.
Is there a target impedance you chase for the CCS at the bottom of a LTPI?
Is there any drawback in using the first node of the bias supply for it? There will be enough negative voltage to use an higher value zener and exceed 1 MOhm with one single bjt.
Are 2 mA fine for the zener?
Thanks in advance.
Is there a target impedance you chase for the CCS at the bottom of a LTPI?
Is there any drawback in using the first node of the bias supply for it? There will be enough negative voltage to use an higher value zener and exceed 1 MOhm with one single bjt.
Are 2 mA fine for the zener?
Thanks in advance.
You'll be doing extremely well if your CCS figure of merit
exceeds 4E+4 ohm*amps
Example: if output impedance is 10 megohms and DC constant current is 5 milliamps, then the figure of merit equals (1E+7) * (5E-3) = 5E+4 ohm*amps. That's excellent performance.
- (Output Impedance) x (DC constant current in amperes)
Example: if output impedance is 10 megohms and DC constant current is 5 milliamps, then the figure of merit equals (1E+7) * (5E-3) = 5E+4 ohm*amps. That's excellent performance.
Thanks @Mark Johnson !
I will use 40 MOhm x mA as a rule of thumb!
So, cascoding two 2N5550 (hfe 150) it will be needed to have a Vb of at least 1.8V, then define the current through the sensing resistor. This will ensure the right impedance vs current.
How to calculate the decrease of impedance vs the frequency?
I will use 40 MOhm x mA as a rule of thumb!
So, cascoding two 2N5550 (hfe 150) it will be needed to have a Vb of at least 1.8V, then define the current through the sensing resistor. This will ensure the right impedance vs current.
How to calculate the decrease of impedance vs the frequency?
This is a really useful rule of thumb I wasn't aware ofYou'll be doing extremely well if your CCS figure of merit
exceeds 4E+4 ohm*amps
- (Output Impedance) x (DC constant current in amperes)
Example: if output impedance is 10 megohms and DC constant current is 5 milliamps, then the figure of merit equals (1E+7) * (5E-3) = 5E+4 ohm*amps. That's excellent performance.
IIRC the best result I had with a simple 2 bjt, or single fet CCS was about 2M at 8mA, (2e^6)*(8e^-3)
So not all that good 🤣
There's an old and very expensive guitar tube amp which used one of these on the bottom of the phase inverter:
https://www.mouser.it/datasheet/2/362/P22-23-CRD-1729293.pdf
Guitar and Bass tube amps are used to have 10k to 22k (with different loads on the two triodes) up to 47k (with equal loads on the two triodes) on the bottom of the phase inverter. So they are designed to be very unperfect.
Are those current regulating diodes in the same range of impedances? I haven't found those information on the datasheet.
https://www.mouser.it/datasheet/2/362/P22-23-CRD-1729293.pdf
Guitar and Bass tube amps are used to have 10k to 22k (with different loads on the two triodes) up to 47k (with equal loads on the two triodes) on the bottom of the phase inverter. So they are designed to be very unperfect.
Are those current regulating diodes in the same range of impedances? I haven't found those information on the datasheet.