Easy alternate FSSA version
Hi FSSA builders
I have been experimenting two very easy modifications on the FSSA original circuit and I believe it can be beneficial for the sound -at least it is in my system. The overall benefit is a more natural sound - to my ear- thus closer to the class A USSA amplifier sound while FSSA being an class AB amplifier.
One of the change appears somewhat counterintuitive since it increases the THD levels but since the FSSA has very low THD levels the results is that the THD levels are still quite low. Also the higher harmonics descending profile is still respected up to clipping.
I have now tested in both FSSA 1 and FSSA 2 versions and it seems to work well. Stay tuned.😉
Fab
Hi FSSA builders
I have been experimenting two very easy modifications on the FSSA original circuit and I believe it can be beneficial for the sound -at least it is in my system. The overall benefit is a more natural sound - to my ear- thus closer to the class A USSA amplifier sound while FSSA being an class AB amplifier.
One of the change appears somewhat counterintuitive since it increases the THD levels but since the FSSA has very low THD levels the results is that the THD levels are still quite low. Also the higher harmonics descending profile is still respected up to clipping.
I have now tested in both FSSA 1 and FSSA 2 versions and it seems to work well. Stay tuned.😉
Fab
Last edited:
This sounds like excellent news Fab!
I haven't built a power amp from scratch since 30y so as stated I will be a very humble and shamefull follower on this project...
I am really eager to see other posts their built to inspire me re housing, power supply components, voltage choice, various cablings and connections, choice of paste for transistors etc. - really starting from scratch but defo willing to learn.
I have the boards, some protective accessories and of course all the transistors. Would like though to order everything else in one go and to venture into known territories re peripherical installation so I can planify fully my assembly, time and planning being key due to my activity...
Guys, please get started to inspire "old new beginners" like me 🙂
Claude
I haven't built a power amp from scratch since 30y so as stated I will be a very humble and shamefull follower on this project...
I am really eager to see other posts their built to inspire me re housing, power supply components, voltage choice, various cablings and connections, choice of paste for transistors etc. - really starting from scratch but defo willing to learn.
I have the boards, some protective accessories and of course all the transistors. Would like though to order everything else in one go and to venture into known territories re peripherical installation so I can planify fully my assembly, time and planning being key due to my activity...
Guys, please get started to inspire "old new beginners" like me 🙂
Claude
Fab,
That's good news. I haven't started my build but probably soon once I get all the parts. Keep us posted.
That's good news. I haven't started my build but probably soon once I get all the parts. Keep us posted.
Thank you both for the cheer up🙂
One change is only the VAS/driver current adjustment with P1/P2 so no BOM part change.
The other is for R40/R41 resistors value. Just have handy 470 or 680 or 750 or 1K or 2k2 ohms.
More precisions to come.
Fab
One change is only the VAS/driver current adjustment with P1/P2 so no BOM part change.
The other is for R40/R41 resistors value. Just have handy 470 or 680 or 750 or 1K or 2k2 ohms.
More precisions to come.
Fab
Last edited:
Hello Fab,
Merry Christmas!
I finished only one channel I'm very satisfied and few days ago I changed the VAS resistors but 330 Ohm value. Is it to low, worthwhile to change again to either? 😕
I would like to ask the size of PCB, positions of holes (if the semiconductors also have I would be grateful) because I want to drill heatsink.
On the first channel the drilling was succesfully, but the second I cocked up and I don't want again 😀.
Thank you in anticipation!
Gy.
Merry Christmas!
I finished only one channel I'm very satisfied and few days ago I changed the VAS resistors but 330 Ohm value. Is it to low, worthwhile to change again to either? 😕
I would like to ask the size of PCB, positions of holes (if the semiconductors also have I would be grateful) because I want to drill heatsink.
On the first channel the drilling was succesfully, but the second I cocked up and I don't want again 😀.
Thank you in anticipation!
Gy.
Hi Fab,
Thank you the fast answers, I will to change 1k.
I didn't know that exist new version of manual, I assembled the amp yet based on rev 0.2.2 (no drill part).
Thanks again
Gy.
Thank you the fast answers, I will to change 1k.
I didn't know that exist new version of manual, I assembled the amp yet based on rev 0.2.2 (no drill part).
Thanks again
Gy.
Yes, locked down at home indeed LOL!
OK... I had hopes being a follower as there are so many things around the excellently documentated Fab amp modules to consider to build a full amp... but indeed not many posters (but you!) to follow LOL!
OK, I have no experience in building a full amp from scratch, and my last build was just boards (the rest was already done) and 3 decades ago. But as no one posts I guess I am not alone with questions, so I decided to tackle that and post my naive (most USSA builders are used to build many amps and have std solutions and know all that by heart I guess) considerations here, to get approval before puchasing expensive parts. Not the boards or the transistors, I have already most of the BOM, I mean here the rest of the hardware: the casing, heatsinks, the PS and all these things that can only be considered in one go and preferably right first time to avoid a mess.
I will post here soon hopefully (I got started) to see if my understanding of all these aspects is correct. I wish I just had to follow and fine tune around an existing base, but no. But then you helped already a lot Anthony... as did Fab and many others. I am just not experimented for such a build and even considered to buld a non expensive chip amp to get rolling before going for the real Mc Coy: this FSSA-2 amp, as I expect a lot of it!
To be followed and hopefully getting a discussion started
Claude
OK... I had hopes being a follower as there are so many things around the excellently documentated Fab amp modules to consider to build a full amp... but indeed not many posters (but you!) to follow LOL!
OK, I have no experience in building a full amp from scratch, and my last build was just boards (the rest was already done) and 3 decades ago. But as no one posts I guess I am not alone with questions, so I decided to tackle that and post my naive (most USSA builders are used to build many amps and have std solutions and know all that by heart I guess) considerations here, to get approval before puchasing expensive parts. Not the boards or the transistors, I have already most of the BOM, I mean here the rest of the hardware: the casing, heatsinks, the PS and all these things that can only be considered in one go and preferably right first time to avoid a mess.
I will post here soon hopefully (I got started) to see if my understanding of all these aspects is correct. I wish I just had to follow and fine tune around an existing base, but no. But then you helped already a lot Anthony... as did Fab and many others. I am just not experimented for such a build and even considered to buld a non expensive chip amp to get rolling before going for the real Mc Coy: this FSSA-2 amp, as I expect a lot of it!
To be followed and hopefully getting a discussion started
Claude
Hey Claude
It's better to layout first ,screw in the boards , then build the power supply, protection, speaker rca,grounding etc. Then u can burn It in while you complete the amp.remember to discharge using a resistor which can take 20mins if crc. It's least frustrating that way. Alot of advanced builder here to comment .
Do amp last ..it will be far neater
The amp is quite difficult. But with Fab around. You won't get into any trouble. The guides is excellent
Get a bigger box ..more room to play you can build even clc + slb or a humongous CLC . And as high possible bias ..or split supply if you plan to do another amp . Oversized trans former have more headroom as well
Do read up bob cordell power supply thread stickied in power supply section .. superb.
It's better to layout first ,screw in the boards , then build the power supply, protection, speaker rca,grounding etc. Then u can burn It in while you complete the amp.remember to discharge using a resistor which can take 20mins if crc. It's least frustrating that way. Alot of advanced builder here to comment .
Do amp last ..it will be far neater
The amp is quite difficult. But with Fab around. You won't get into any trouble. The guides is excellent
Get a bigger box ..more room to play you can build even clc + slb or a humongous CLC . And as high possible bias ..or split supply if you plan to do another amp . Oversized trans former have more headroom as well
Do read up bob cordell power supply thread stickied in power supply section .. superb.
Last edited:
Oh yeah ..
The power supply quality will be the determining factor between great to super amp .. parallels multi caps to lower impedance.mkt film cap bypass .good quality oversized transformer choke....that what factory find expensive to make ..we can go crazy ..no need expensive caps ..just good quality ..
I build slb type before ..but a great clc is shocking good .
Just my experience
The power supply quality will be the determining factor between great to super amp .. parallels multi caps to lower impedance.mkt film cap bypass .good quality oversized transformer choke....that what factory find expensive to make ..we can go crazy ..no need expensive caps ..just good quality ..
I build slb type before ..but a great clc is shocking good .
Just my experience
Hi Anthony,
I am with you and in fact I read so far 4 books on amps. All didn't converge though, so as I am an old engineer I am also trying to make my selection. The downside is I never ever had to bolt a transistor on a heatsink, so I am really starting from scratch from plug / switch to bridge and pratcila considerations and of course your advices re slow start and DC protection...
The thing is I can't really start on a bench as I don't have the space for that, nor an existing PS that can cope (max 30V and 2.5A). So I need to plan ahead to get it right first time, and that means starting at the PS and grounding sadly. I am completely with you that PS is the key, once the circuit has been designed. Assembling the boards themwelves are not the problem, it is what they bolt on that is.
The positive is the great support and indeed the amp board itself which is very well documented, so that I shall have so far only 2 component questions regarding these. BUT so many options and questions for ALL the rest!
To be followed very soon
Claude
I am with you and in fact I read so far 4 books on amps. All didn't converge though, so as I am an old engineer I am also trying to make my selection. The downside is I never ever had to bolt a transistor on a heatsink, so I am really starting from scratch from plug / switch to bridge and pratcila considerations and of course your advices re slow start and DC protection...
The thing is I can't really start on a bench as I don't have the space for that, nor an existing PS that can cope (max 30V and 2.5A). So I need to plan ahead to get it right first time, and that means starting at the PS and grounding sadly. I am completely with you that PS is the key, once the circuit has been designed. Assembling the boards themwelves are not the problem, it is what they bolt on that is.
The positive is the great support and indeed the amp board itself which is very well documented, so that I shall have so far only 2 component questions regarding these. BUT so many options and questions for ALL the rest!
To be followed very soon
Claude
So one needs to get started somewhere. It is a bit like chicken and egg designing an amp. On one side you need to know what entry you have (read PS, voltage etc.), on the other side you can't define that if you don't decide on the output (delivery, LS needs). But you can't decide on the output without considering PS and heatsinks etc. which are the input
Argh!!
So, as for every project, I decided to start with the goal and see how to get there.
I long time hesitated betwen the FFSA-2 and the USSA amps. I don't need a lot of power, in fact when I run my LS I usualy need a few Watts. They are advertised and measured by the press at 87dB/W/m (they are rate 4R so that's 2.8V) and impedance can go down to 2.6R worst case with moderate phase rotation. It could probably do well with strong 40W, but I wanted some headroom in case of, and also a good damping factor. Therefore I decided to go for the former, assuming Fab would squeeze most of the sound excellency he is used to while Anthony confirmed it was an amp with grip on LS, which I am after. In short, I hope to get anything between 5 and 10W under 4R in Class A, while not being down on max power I have today, in case of, and that's roughly 150W into 4R.
I don't need more power as running a lower voltage enables me more bias with the same heatsinks, and I am more after that. I need an amp that can deliver stable power in the 2-4R range.
Say I want say 150W at 4R (could survive with 10% less but that's the starting point and right ballpark). My current amp can deliver 200W in 3R on impulse and is 2R stable with 220W or so. I want that low impedance capacity aswell. I also want as much Class A as I can into 4R (= high bias) because that's where my speakers will play the most.
Whereas I am used on my small devices to go for ultra fast soft diodes, I will follow Papa's advice re diode bridge / rectifier for amps and that means a bog std diode bridge with 0.1 caps between each connection to tame it. That means std losses. I don't care for losses TBH given what they are. I do care for reliability. Never ever seen a rectifier die.
I worked out that going for a transformer rated for 30V output I would get roughly 42V DC at the entry of board's power rails (neglecting for the moment the losses by say a CRC filter) and that should hopefully translate into 150W at 4R while also enabling the safe use of caps that are rated for 50V. I took for this the std losses I found in the litterature, I am well aware that the LS won't see + and - 42V, but I figured out that 42V would translate into perhaps more something like "at least 34.6V SIN capability, eg 24.5V avg" at the output, hence 150W into 4R.
Is this correct?
If so, I would have thought that this would translate, provided the transistors can deliver that (they may not on specsheet, or on impulse only?) into 300W into 2R (quite academic)) and more importantly and realisticaly max 200W into 3R.
Am I right? I am somewhat confused by what the manual says, eg that going from 8R to 4R you don't double the max power output but have roughly 1.7 times the power (provided the PS can cope). I must confess I don't understand why...
If all this is correct then I am about to release a PS schematic from the main's plug to +V and -V to feed the amp module. I like that because the litterature mainly just considers the transformers and filtering, not showing the full picture. That will for sure open a lot of discussions, and hopefully be of some use for all the builders that want to get started but don't where to start.
But let's first get the basics right, eg these figures...
Many thanks for your kind help
Claude
Argh!!
So, as for every project, I decided to start with the goal and see how to get there.
I long time hesitated betwen the FFSA-2 and the USSA amps. I don't need a lot of power, in fact when I run my LS I usualy need a few Watts. They are advertised and measured by the press at 87dB/W/m (they are rate 4R so that's 2.8V) and impedance can go down to 2.6R worst case with moderate phase rotation. It could probably do well with strong 40W, but I wanted some headroom in case of, and also a good damping factor. Therefore I decided to go for the former, assuming Fab would squeeze most of the sound excellency he is used to while Anthony confirmed it was an amp with grip on LS, which I am after. In short, I hope to get anything between 5 and 10W under 4R in Class A, while not being down on max power I have today, in case of, and that's roughly 150W into 4R.
I don't need more power as running a lower voltage enables me more bias with the same heatsinks, and I am more after that. I need an amp that can deliver stable power in the 2-4R range.
Say I want say 150W at 4R (could survive with 10% less but that's the starting point and right ballpark). My current amp can deliver 200W in 3R on impulse and is 2R stable with 220W or so. I want that low impedance capacity aswell. I also want as much Class A as I can into 4R (= high bias) because that's where my speakers will play the most.
Whereas I am used on my small devices to go for ultra fast soft diodes, I will follow Papa's advice re diode bridge / rectifier for amps and that means a bog std diode bridge with 0.1 caps between each connection to tame it. That means std losses. I don't care for losses TBH given what they are. I do care for reliability. Never ever seen a rectifier die.
I worked out that going for a transformer rated for 30V output I would get roughly 42V DC at the entry of board's power rails (neglecting for the moment the losses by say a CRC filter) and that should hopefully translate into 150W at 4R while also enabling the safe use of caps that are rated for 50V. I took for this the std losses I found in the litterature, I am well aware that the LS won't see + and - 42V, but I figured out that 42V would translate into perhaps more something like "at least 34.6V SIN capability, eg 24.5V avg" at the output, hence 150W into 4R.
Is this correct?
If so, I would have thought that this would translate, provided the transistors can deliver that (they may not on specsheet, or on impulse only?) into 300W into 2R (quite academic)) and more importantly and realisticaly max 200W into 3R.
Am I right? I am somewhat confused by what the manual says, eg that going from 8R to 4R you don't double the max power output but have roughly 1.7 times the power (provided the PS can cope). I must confess I don't understand why...
If all this is correct then I am about to release a PS schematic from the main's plug to +V and -V to feed the amp module. I like that because the litterature mainly just considers the transformers and filtering, not showing the full picture. That will for sure open a lot of discussions, and hopefully be of some use for all the builders that want to get started but don't where to start.
But let's first get the basics right, eg these figures...
Many thanks for your kind help
Claude
Didn't state the obvious, but I am of course refering to a 2x 30V tranny to get +V and -V, only gave values for one rail... mail being long enough, sorry 🙂
Hi Claude
Your numbers for PSU voltage roughly matches page 4 of the FSSA build manual when using a standard bridge loss (which is fine). However because of the cap multiplier loss on the pcb it gives more about 130Wrms into 4 ohms with +/-42VDC. For the loss in the R of CRC filter, the R of about 0.1 to 0.15 ohms will make a small voltage drop even at 0.7A bias.
The big decision for you is to decide if you go with a main PSU for the whole amp or use an additional separate PSU with higher voltage for the front end of the amplifier. See chapter 7 of the FSSA manual. If you use 2 separate PSU voltages then this will increase the max power of the amp. For example, +/-42VDC main PSU with +/-49VDC secondary PSU for the front end will give you roughly 150Wrms into 4 ohms.
As for the tranny, I would use 2 x 32VAC to get +/-42 to 44 VDC (depending on your min and max line voltage ) with reasonable transfo rating and PSU caps with standard rectifier bridge. These are approximate numbers and usung one or two bridges per polarity will also change slightly the result.
But really 130W or 150W gives a very small difference in Sound pressure level. Remember the general formula P = V*V / R.
Fab
Your numbers for PSU voltage roughly matches page 4 of the FSSA build manual when using a standard bridge loss (which is fine). However because of the cap multiplier loss on the pcb it gives more about 130Wrms into 4 ohms with +/-42VDC. For the loss in the R of CRC filter, the R of about 0.1 to 0.15 ohms will make a small voltage drop even at 0.7A bias.
The big decision for you is to decide if you go with a main PSU for the whole amp or use an additional separate PSU with higher voltage for the front end of the amplifier. See chapter 7 of the FSSA manual. If you use 2 separate PSU voltages then this will increase the max power of the amp. For example, +/-42VDC main PSU with +/-49VDC secondary PSU for the front end will give you roughly 150Wrms into 4 ohms.
As for the tranny, I would use 2 x 32VAC to get +/-42 to 44 VDC (depending on your min and max line voltage ) with reasonable transfo rating and PSU caps with standard rectifier bridge. These are approximate numbers and usung one or two bridges per polarity will also change slightly the result.
But really 130W or 150W gives a very small difference in Sound pressure level. Remember the general formula P = V*V / R.
Fab
Last edited:
Hi Fab
Many many thanks for your kind answer, spot on!
I will already correct the schematic I am drawing with the right vaues, I am glad I asked... and you answered.
I still don't know where that 1.7 factor instead of 2 came when going from 8R to 4R, but here is what I now intend to do :
- I will go for 2x 32VAC transformers instead of 2x30V, to make sure I get "at least 42V" line voltage. I undertsand I won't get as much as 150W, but I will get "at least 130W under 4R" and probably more like 140W. I can live with that...
- I will go with a main PSU for the whole amp. I don't need more power under 4R, more interested into power and phase stability at 2R.
Many thanks again Fab!
Claude
Many many thanks for your kind answer, spot on!
I will already correct the schematic I am drawing with the right vaues, I am glad I asked... and you answered.
I still don't know where that 1.7 factor instead of 2 came when going from 8R to 4R, but here is what I now intend to do :
- I will go for 2x 32VAC transformers instead of 2x30V, to make sure I get "at least 42V" line voltage. I undertsand I won't get as much as 150W, but I will get "at least 130W under 4R" and probably more like 140W. I can live with that...
- I will go with a main PSU for the whole amp. I don't need more power under 4R, more interested into power and phase stability at 2R.
Many thanks again Fab!
Claude
Hi Claude
The 1.7 factor for 8 ohms to 4 ohms is only physics here of using real intended components and not ideal components. There are several reasons why the factor is not 2. Some of them are PSU voltage load regulation that is reduced at higher load current (see transfo datasheet), rectifier diode bridge voltage drop increase at higher load current, RDSon of mosfet that has more impact on losses at high current, etc....
When you see an amplifier specifications giving a factor of 2 you can assume that they are not giving the max power value for higher load speaker value. For example,
1)
100Wrms into 8 ohms
200Wrms into 4 ohms
2)
Could be interpreted - for example- as:
117Wrms into 8 ohms
200Wrms into 4 ohms
So spec of 1 is still true even if it spec 2 in reality.
The ratio is mostly dependent of power supply sizing and after that amplifier topology and capability used.
Fab
The 1.7 factor for 8 ohms to 4 ohms is only physics here of using real intended components and not ideal components. There are several reasons why the factor is not 2. Some of them are PSU voltage load regulation that is reduced at higher load current (see transfo datasheet), rectifier diode bridge voltage drop increase at higher load current, RDSon of mosfet that has more impact on losses at high current, etc....
When you see an amplifier specifications giving a factor of 2 you can assume that they are not giving the max power value for higher load speaker value. For example,
1)
100Wrms into 8 ohms
200Wrms into 4 ohms
2)
Could be interpreted - for example- as:
117Wrms into 8 ohms
200Wrms into 4 ohms
So spec of 1 is still true even if it spec 2 in reality.
The ratio is mostly dependent of power supply sizing and after that amplifier topology and capability used.
Fab
Last edited: