I noticed in your sony 40 year anniversary vfet amp you used TO220 bipolar devices as cascodes.
Is this because they are generally superior to the TO92 bipolars or is there a dumber reason (eg you already have a truck load of them that are begging to be used in something)
The reason I ask is today I came across a box containing a stack of 2sc5171/2sa1930 and 2sc4793/2sa1837, they are not going to be used in anything else so if these are great devices for cascodes (ie better than TO92), then I can finally put them to use.
Feel free to say anything else on this as I'm sure to learn something new.
Is this because they are generally superior to the TO92 bipolars or is there a dumber reason (eg you already have a truck load of them that are begging to be used in something)
The reason I ask is today I came across a box containing a stack of 2sc5171/2sa1930 and 2sc4793/2sa1837, they are not going to be used in anything else so if these are great devices for cascodes (ie better than TO92), then I can finally put them to use.
Feel free to say anything else on this as I'm sure to learn something new.
Last edited:
I believe the 2sc4793/2sa1837 were Papa's choice for cascodes in the F5 turbo article. I wil be using them in a cascode for the BA3 preamp I have to increase the voltage to 36V. (straighforward copy of the f5turbo cascode section)
They have higher dissipation values than the to92 this might be a factor in some designs such as the F5turbo depending on rail voltages. I don't remember the rail voltage in the sony amplifier but if these must drop lots of voltage the To92 parts might be too small.
And he probably has a truckload of those... 😛
Question: Can you just lift the cascade circuit off one amp and install it on another? Excuse my lack of knowledge here, but I thought resistors and so forth had to be calculated for each application. I will be happy to be wrong here!
Russellc
Pico - I bet Papa's answer will be his normal, "they are of proper dissipation and I had 10,000 of them" 😀 😀 😀
Russell - Basically yes. In theses amps the cascode usually sits around 1/2 the rail voltage, to protect the Jfet from seeing big rails in a higher-voltage amp. Look at the F5Tv3 for a good place to start.
Russell - Basically yes. In theses amps the cascode usually sits around 1/2 the rail voltage, to protect the Jfet from seeing big rails in a higher-voltage amp. Look at the F5Tv3 for a good place to start.
well , decent Hfe and not too big capacitances , are two things worth caring
for EU boyz , if obtaining decent BD139/140 is possible , they're good and cheap solution
for EU boyz , if obtaining decent BD139/140 is possible , they're good and cheap solution
I got the idea from a member who built an awesome BA3 power amp using 36V rails.
He used the F5 turbo cascode circuit with good results. The input stage of the BA3 and F5 turbo is very similar and in this case the voltages are similar so i can get away with the same resistor values. If you change the voltages it is just a matter of adjusting the resistors to make sure you have the right voltage going into the jfets.
BTW I have been running my BA3 input jfets at 30V for the last 4 months with no problems, they are at 40-44º C. 🙂
My understanding is, hearing this from the partners at Pass Labs, his normal conversation goes something like, "So how many of them do I need to buy to get the price I want?"
And this is why I saw eye-watering amounts of long out of production amazing transistors on the shelves in their production storeroom.
Very cool.
And this is why I saw eye-watering amounts of long out of production amazing transistors on the shelves in their production storeroom.
Very cool.
Anything around 36V to 40V would in theory still be ok in most situations but I suppose if Papa prefers a lower Vds value on the jfets say 12V Vds, then yeah a baby bipolar to92 is not going to last as long as a to220.
Same here, complimentary BA3 with 31 volt rails. This amp has been in service a while, no probs. I had thought of cascading it, run the jfets a little cooler, and isn't there also a slight improvement in distortion?
Thanks, and 6l6 too!
Russellc
Last edited:
In theory the cascode shields the input jfets from some of the noise in the psu rails, so it could help with measurements. I am running a crippled F4 (same as BA3 OS) on +-33V PSU (it drops to 30V when heavily biased, around 400mA per transistor) and I want the front end to have some extra voltage swing to make sure the OS is driven to its full potential.
The reason I tend to favor the bipolars as front-end cascodes is that they
tend to have fewer problems with parasitic oscillation, requiring fewer
parts for stable operation. (By this I mean a resistor....)
The cascode transistor itself having very little contribution to the character,
there's no problem with decent bipolars. It's just a voltage umbrella for
the more delicate gain part.
tend to have fewer problems with parasitic oscillation, requiring fewer
parts for stable operation. (By this I mean a resistor....)
The cascode transistor itself having very little contribution to the character,
there's no problem with decent bipolars. It's just a voltage umbrella for
the more delicate gain part.
Sorry Nelson, just so I can sleep better at night.........
The question was, why do you prefer TO220 bipolar devices over TO92 bipolar devices?
I suspect the answer is probably insignificant, but this is a matter of sleeping better at night.
Thank you. 🙂
Last edited:
The dissipation is also needed in some of the circuits, we have used TO92 but the TO220 performs better.
- Status
- Not open for further replies.