folded cascode puzzling

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.

I'm currently trying to understand more deeply some common configurations to
this forum such as the folded cascode stage employed in the X amplifier series...

To me this folded cascode thing seems to serve multiple purposes and the more I look at it the more that I think that such configuration is at the same time a cascode stage, a constant current source for the front end, a shift voltage level, etc.....

My questions are:

Is this configuration (folded cascode output) able to provide some sort of current gain?

Also, which is the best current ratio for the front end and cascode output?

-The very first thing that occurs to me is that the front end and the folded
cascode output should share the same current amount-

Can I assert if I say that this folded cascode stage also provides some sort of voltage amplification?

I hope you can help to disipate my 'clouds'.....

Folded Cascode

First do some research on how a basic cascode circuit works. Nelson Pass wrote an article at:

The folded cascode works in basically the same manner as the regular cascode. The signal current from the gain transistor is the same the same AC current as in the cascde transistor. Since the folded cascode has an addtional current souce the DC currents for the gain and cascode transistor can be different. This is where the real usefulness of this circuit comes to play. The folded cascode does provide some advantage over a standard cascode.

A smaller power supply voltage is required for a given voltage gain.

Greater power supply rejection. The load resistor uses ground as reference in stead of the power supply rail. The load on the power supply is also a constant current instead of the signal current.

Greater flexibility in terms of the load resistor. Since the DC bais current of gain transistor does not have to appear across the load resistor it can be larger for a give supply voltage and greater gain can be achieved.

Re: folded cascode....


Many thanks for your help I'll be 'ruminating' what you wrote for a while...


Did you actually designed the LM6313?, To me it is a cleaver design and some
time ago I was toying with the idea of making a more 'powerful' version of it,
let say in the range of 15W @ 8 ohms ....

About LM6313, still I have to do more 'thinking' about the workings of the
current mirror that feeds the cascode stages and polarizes the output pair...

Why I want to connect a capacitor on pin 15? ( Is it to provide instantaneous
power for the output pair when they are doing current peaking?...

Also, I'm still thinking about the possibility of take out the output pair of
the voltage section an connect directly the cascoded+current mirror directly
to the current amplifier section, but still I don't get it right...

Well, returning to the folded stage (And to my self imposed challenge,
'produce a decent amplifier using common man available parts')...

Once I read (patent 4,406,990) that the current ratio for the folded cascode
versus the front end should be half of the later in order to improve noise and
frequency specs but nobody else (that I'm aware off) seems to care about it.

In the other hand, I'm in the process of re-take an old project (15Wrms @ 8 Ohms,
30Wrms @ 4 Ohms) based on the X topology but using a lot of my college spare
parts and that involves a lot of 2N5457's, 2N5460's, MTP3055's, 2N2222's, and
2N2907's. I already did the matching and parts selection. Since I don't have
P-Mosfets, I will use a X front end attached to an Aleph output and here is
where I'm running in trouble.....

For once, the 2N5457 & 2N5460 handle low voltage and very low current for my
taste so I thought about paralleling some of them until I can get a decent
amount of current to drive the MTP3055 input capacitance. Also I have been toying
with the idea of using the 2N2907 to make the folded cascode and save some trouble
but then I should run (whatever I select for the folded cascode) to a higher
current than the front end and this is the point where I wonder how it will affect
the amplifier since this is opposite to what is said on that document (patent
4,406,990)... is just that another famous audio designer buddy asked me if I got he idea for the amp I designed for XYZ Co. was ripped off from that IC.

Actually, I got the idea for the output stage from an old Audio Research D-100........from the 70's. I think I stole the folded cascode from the AD797.....or one of those designs.

Let me peruse the rest before I respond........

Looking at the '6313 wonder it makes little sense. Very poorly drawn.

Pin 15 is for compensation. Move the cap to either side of those diodes,and it should make better sense.

Try redrawing the '6313 with all the transistors going from top to bottom should help too. Those 2 going horizontal make it hard to read. Draw the diodes in series with the current-mirror, and then draw the followers separate, and just connect the bases.

Current ratio? Just because it is it the patent application, doesn't meant they are actually doing it. I bet Nelson knows about that........[joke].

I'm not familiar with the Pass circuit.....I bet Harry is, but if it has a MOSFET you need to drive as the "A2" of the '6313. you have to be able to drive the capacitance. You need a decent buffer to drive a MOSFET. Don't drive it from the voltage gain stage directly.

As for the parts you describe........yuck! Didn't you read my "new law" a few weeks ago?

Any more questions?

Re: new law.....


Thanks for your response... I still have to redraw the LM6313 to understand it

I absolutely agree with you, my part selection is a real poor choice... I also
read your post (new law) and I agreed with it.... but I could not refrain some
ironic smile... partly because I had such parts, partly because I live in a third
world country and some times is hard to get these more modern parts in small
amounts locally (besides those all not so good parts)... enough complaining!...

Maybe I'll end not using these parts at all, but I have more than 50 of each

Well, after some digesting from my part I'll get back with more questions
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.