and this one by Hitsware: http://www.diyaudio.com/forums/solid-state/189766-square-law-class-amps-2.html#post2956849
Hello forr and vzaichenko.
This simplified schematic is for a voltage source amp. It can be converted to a schematic of a current source amp as follows:
1. Insert a 1 Ohm non-inductive power resistor [for example] between the joined sources of the power output Mosfets and ground. The voltage gain of this common source stage is equal to 8.
2. The joined Mosfets' sources now becomes the new point of origin for the overall negative feedback [ONF] loop.
3. Flip the inputs of the front end Op Amp; meaning the music signal is now presented to the non-inverting port instead of the inverting input.
4. The ONF circuit [as shown in the simplified schematic] is then connected to the inverting port of the op amp as commonly done.
5. The current flowing out of the Mosfet sources is cleaned up by ONF. It is also equal in magnitude to their drains' current.
6. It follows that the distortions are similarly minimized in the output drain current flowing through the 8 Ohm load.
7. The overall gain of this amp is a whopping [28 x 8] =224. May need to lower the 28 part which is [1+27K/1K] to suit.
8 . Point 7 also says that that the trans-conductance [Gm] of this amp = 224 Amps out per 1 input Volt.
Hi Guys
Valery: Even though the current path for the output stage is identical for the grounded supply and floating supply amplifiers, they do not offer the same performance. The key is because of the separation of the "point of regulation" and the "point of feedback". This is very obvious in a tube amp where the output transformer has impedance taps but feedback is permanently tied to only one.
In the solid-state amps in question, there are many examples of the floating output stage and none have good performance. The advantage of limiting the need for high-voltage parts to just the output stage is insignificant with today's devices inasmuch as one can find suitable devices for any and all stages of the amplifier at low cost. One can have an output stage with gain which satisfies the desire for limited need of high-voltage devices, or one can use floating drive circuits where only low-voltage devices or opamps are needed. The former approach can provide stellar performance where the latter is troubled.
The circuit linked from post-101 won't work. Both pairs of mosfet are turned on/off identically so there will be no signal across the load.
Have fun
Valery: Even though the current path for the output stage is identical for the grounded supply and floating supply amplifiers, they do not offer the same performance. The key is because of the separation of the "point of regulation" and the "point of feedback". This is very obvious in a tube amp where the output transformer has impedance taps but feedback is permanently tied to only one.
In the solid-state amps in question, there are many examples of the floating output stage and none have good performance. The advantage of limiting the need for high-voltage parts to just the output stage is insignificant with today's devices inasmuch as one can find suitable devices for any and all stages of the amplifier at low cost. One can have an output stage with gain which satisfies the desire for limited need of high-voltage devices, or one can use floating drive circuits where only low-voltage devices or opamps are needed. The former approach can provide stellar performance where the latter is troubled.
The circuit linked from post-101 won't work. Both pairs of mosfet are turned on/off identically so there will be no signal across the load.
Have fun
The key thing is that in both cases NFB is taken from the load, so that all kinds of "error" signals will be taken care of. There are many other ways of limiting the need of high-voltage parts - I often use +/-15V rails for the front-end in my designs, switching to full swing of +/-70V right in front of the drivers - that's no problem.
I'm still pretty sure that if a good PSU is used, OPS with gain and floating central tap does not suffer from any disadvantages, more serious than if it would be a traditional arrangement.
Schematic from post #101 - if you look carefully, the first pair is common source, the second one is common drain (follower). They form a bridge - well, rather non-standard topology - but there will be a signal across the load for sure.
Hafler 9505 Power Supply Capacitor Question
Hello,
I have a Hafler 9505 trans-nova amplifier and it needs the power supply capacitors but I cant seem to find them anywhere!!! The amp does not have the capacitors installed so I don’t know of the original part number but it needs to be 2.5” diameter and no taller than 3.625” tall with 20,000 uF @ 100 volts.
I have a 85volt capacitor in it now but am not sure if it is safe!!! If you can give me a supplier that will sell me 4 caps I would really appreciate it. I have 10 Hafler amplifiers and counting !!!
Hello,
I have a Hafler 9505 trans-nova amplifier and it needs the power supply capacitors but I cant seem to find them anywhere!!! The amp does not have the capacitors installed so I don’t know of the original part number but it needs to be 2.5” diameter and no taller than 3.625” tall with 20,000 uF @ 100 volts.
I have a 85volt capacitor in it now but am not sure if it is safe!!! If you can give me a supplier that will sell me 4 caps I would really appreciate it. I have 10 Hafler amplifiers and counting !!!
sorry, no easy solution for those; they indeed are tough to find.
new, known good parts (i.e. non-ebay) probably won't be cheap.
start with a 20,000uF 100v in google and work down the list until you get a hit.
if it helps, probably ok to use from 15,000uF to 25,000uF (or even wider) to open your choices a bit.
those big electrolytics often have wide tolerances
good luck; it's a nice amp.
mlloyd1
new, known good parts (i.e. non-ebay) probably won't be cheap.
start with a 20,000uF 100v in google and work down the list until you get a hit.
if it helps, probably ok to use from 15,000uF to 25,000uF (or even wider) to open your choices a bit.
those big electrolytics often have wide tolerances
good luck; it's a nice amp.
mlloyd1
Hello,
I purchased the amplifier without the caps and covers so I was able to locate these caps Cornell Dubilier 23000UF 85V Electrolytic Capacitor DCMX233M085CB2B DCMC and they work but I am not sure if I push the amplifier they will explode so what are you thoughts on this and what is limited lifetime mean, 1 year, 3 years... I can get the caps but they are to tall and I will be adding custom covers.
Thanks and any information will help!!!
I purchased the amplifier without the caps and covers so I was able to locate these caps Cornell Dubilier 23000UF 85V Electrolytic Capacitor DCMX233M085CB2B DCMC and they work but I am not sure if I push the amplifier they will explode so what are you thoughts on this and what is limited lifetime mean, 1 year, 3 years... I can get the caps but they are to tall and I will be adding custom covers.
Thanks and any information will help!!!
As mentioned in previous posts on this subject, 85-volt capacitors offer an insufficient safety margin. They should be rated for 100-volts like the originals. This is not just a case when you "push" the amplifier, in fact, just the opposite. At idle, the voltage on the caps will be at maximum, and if your line voltage is running a little high, the caps could fail in a spectacular fashion. At best, 85-volts caps will not last very long in this application
The sad news is that the original caps were custom made for this amplifier, and after doing exhaustive research for equal replacements (i.e. the same electrical and dimensional ratings) I have concluded that no one currently makes a suitable off-the-shelf part. The only parts that I have found with the proper electrical ratings, the correct terminal spacing for mounting to the board, and a sufficiently small diameter, are too tall to fit in the amplifier. So the only "simple" choice is to remove the cover entirely (not a great idea) or cut large holes in the cover to clear the too-tall capacitors.
All of this is a shame, as these are great amplifiers, and otherwise pretty reliable, but due to their increasing age, capacitor failures are becoming at least somewhat common. I'm crossing my fingers that my own "engineering prototype" 9500 hangs in there, as it's been in continuous use since about 1990 and I do not want to part with it!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.