Nelson,
I can't thank you enough for your help. I'm having a blast trying all these amps.
Are you saying you think I'll end up with the Balanced-Cascoded F5, or just straight Cascoded F5?
Have you tried a Balanced-Cascoded version of the F5 in your testing? It's been suggested that cascoding does not add to the balanced version, has that been your experience?
Finally, if I read your PS post in the other thread correctly, you think a 12V rails biased as high as I can take it will be the way to go for the Balanced version, and ~22V rails for the Cascoded version (you were using 50W peak as an example). I haven't seen an amp with 12V rails, am I reading your post correctly?
"The journey is the thing." and "Just build it!"
I can't thank you enough for your help. I'm having a blast trying all these amps.
Are you saying you think I'll end up with the Balanced-Cascoded F5, or just straight Cascoded F5?
Have you tried a Balanced-Cascoded version of the F5 in your testing? It's been suggested that cascoding does not add to the balanced version, has that been your experience?
Finally, if I read your PS post in the other thread correctly, you think a 12V rails biased as high as I can take it will be the way to go for the Balanced version, and ~22V rails for the Cascoded version (you were using 50W peak as an example). I haven't seen an amp with 12V rails, am I reading your post correctly?
"The journey is the thing." and "Just build it!"
I have pretty much gone through the permutations listed,
and it's a worthy progression.
In the end, it is probable that you will settle on Cascoded
2SJ109's and 2Sk389's driving Toshiba's biased as high
as you dare. Get yourself lots of heat sinking and power
supply capacitors, and use bipolars for the cascodes.
In this circuit, balanced mostly brings more power to the
table.
Minimize the feedback where you can.
The journey's the thing. If you just skip to that, you'll
miss out on some cool stuff.
Sorry for the confusion.
http://www.diyaudio.com/forums/swap-meet/179514-huge-caps-nippon-chemi-con-68000uf-75vdc.html
Contact user UDailey. He is selling 68000Uf 75V caps.
http://www.diyaudio.com/forums/swap-meet/179514-huge-caps-nippon-chemi-con-68000uf-75vdc.html
Contact user UDailey. He is selling 68000Uf 75V caps.
Last edited:
The fets like to have some extra voltage, as their
capacitance gets high at low Vds, so I always keep
some leftover voltage.
Cascoding basically improves the performance of most
parts operated in the saturated region, and some in the
"linear/ohmic" region as well. It's a matter of how much
and whether you want the additional complexity in the
circuit. Tempting DIYers to build an amplifier is best
done with very simple circuits.
In the end, it is probable that you will settle on cascoded
2SJ109's and 2Sk389's driving Toshiba's biased as high
as you dare. Get yourself lots of heat sinking and power
supply capacitors, and use bipolars for the cascodes.
In this circuit, balanced mostly brings more power
I think I am correct in assuming that you are saying this about the cascoded "standard" F-5, and that the balanced version just brings more power? In other words, cascoding, and biasing "as high as you dare"? If so, will this work with standard voltage for the F-5, or should this also be ramped up? Asking because I'm getting ready to purchase transformers and heat sink for the next few projects.
Thanks for your time. (I suppose there is such thing as too many 18+18 here!)
Russellc
Last edited:
its in the logic my friend
its been said very clearly, mosfets excels at higher voltage, and higher bias
going balanced with this amp is quite the opposite
but what you do get is more power
right now I wonder about the cascoded Jfets
means that you can run it at 30V rails
but the Jfet and cascode shares this voltage
does this mean they only get 15Vdc each
15V is a bit low, no ?
right or wrong ?
its been said very clearly, mosfets excels at higher voltage, and higher bias
going balanced with this amp is quite the opposite
but what you do get is more power
right now I wonder about the cascoded Jfets
means that you can run it at 30V rails
but the Jfet and cascode shares this voltage
does this mean they only get 15Vdc each
15V is a bit low, no ?
right or wrong ?
No.
It's more than enough. 10V is the usual Vds for JFET in cascode arrangement. E.Borbely didn't gave them more than 2-3V.
Actual voltage swing of F5's input stage is couple of volts max.
I know it is nice to get a response from Nelson, but these questions have been answered more than a few times now.
Either you are not reading peoples responses to your questions or you don't understand or believe the response.
If you don't understand the response then say so, otherwise people will lose interest in answering any of your questions, if you constantly ignore them.
Maybe I missed what you're talking about, I don't read all the threads. I think I understand the answers, and I think I read the threads I'm in.
In the last Balanced thread, it was suggested by two members that Cascoding did not help with the Balanced F5. Nelson suggested it did, thus the question.
It seems like you may have missed a post or two though:
http://www.diyaudio.com/forums/pass-labs/172770-balanced-f5-question-62.html#post2371824
Did I miss read the two fellows from the other thread, or did I misinterpret Nelson's suggestion?
Finally, I don't think anyone has Nelson's experience with the F5, so if he's willing to opine, I'll take his response 100% of the time over other other "experts".
In the last Balanced thread, it was suggested by two members that Cascoding did not help with the Balanced F5. Nelson suggested it did, thus the question.
It seems like you may have missed a post or two though:
http://www.diyaudio.com/forums/pass-labs/172770-balanced-f5-question-62.html#post2371824
Did I miss read the two fellows from the other thread, or did I misinterpret Nelson's suggestion?
Finally, I don't think anyone has Nelson's experience with the F5, so if he's willing to opine, I'll take his response 100% of the time over other other "experts".
Last edited:
I prefer higher voltages due to the increased capacitance
of the devices at low voltages. Any circuit output
approaching the rails will experience this in any case.
The Toshibas are very nice parts. Since they have been
discontinued, I only have enough for some prototype
work out of interest.
Cascoding the inputs is always useful. Cascoding the
outputs chews up voltage, so you will always need higher
rails to do that.
Just in case you did not get the message. You want high voltage for the reasons stated above, therefore cascoding on the front end becomes mandatory. I would not push the Jfets much past 30V, for two reasons.
1. It will limit the amount of current you can put through them to drive the output stage.
2. The Jfets are not rated for high voltage (spec sheet says 25V)
You want high current on the output for reasons you already understand.
This means monoblocks. Buy another couple of those conrad heatsinks and you have the freedom to build what ever you want within reason.
All of this will depend on the type of speakers you have and how loud you listen to music.
If you have 100dB speaker and listen at quiet levels then you probably don't need to go that extreme and the stock F5 or Patrick's F5X will more than do the job with excellent results.
However I am guessing your speakers are not 100dB and that you might listen at elevated levels when the wife is out
1. It will limit the amount of current you can put through them to drive the output stage.
2. The Jfets are not rated for high voltage (spec sheet says 25V)
You want high current on the output for reasons you already understand.
This means monoblocks. Buy another couple of those conrad heatsinks and you have the freedom to build what ever you want within reason.
All of this will depend on the type of speakers you have and how loud you listen to music.
If you have 100dB speaker and listen at quiet levels then you probably don't need to go that extreme and the stock F5 or Patrick's F5X will more than do the job with excellent results.
However I am guessing your speakers are not 100dB and that you might listen at elevated levels when the wife is out
i'm biamping my speakers with ajx for bass and f5x for highs. speakers are only 2 ways and the woofers are 8 inch scanspeak 21w/8554 8 ohms (90db) and tweeters are 6 ohms scanspeak d2904/9700 (89.5db). i do not wish to go with monoblocks and am using the same heatsinks as nycone, one case for each highs and lows. i have a small room so they will only very rarely be pushed and was planning to lower the rails and crank up the bias, i will use a fan if need be rather than building 4 amps
does this splitting up of duties for a pair of stereo balanced amps make for effectively the same as going monoblocks with one for each speaker for stereo balanced with passive xo?. my math seems to indicate i wont have any problem going louder than i will ever need to without leaving class A, but this is my first speaker amp and speaker build, so i'm open to other opinions.
does this splitting up of duties for a pair of stereo balanced amps make for effectively the same as going monoblocks with one for each speaker for stereo balanced with passive xo?. my math seems to indicate i wont have any problem going louder than i will ever need to without leaving class A, but this is my first speaker amp and speaker build, so i'm open to other opinions.
It goes fast with a handheld flexible dremel (10 min). With a drill press about 15 min.
You could go surface mounted - I'm going to try that after I settle on an amp. For now - drill holes - it's only 15 min compared to years of knowing you did the whole amp yourself!
Now that I've etched PCB, I'm going to have a hard time buying commercial PCB.
You could go surface mounted - I'm going to try that after I settle on an amp. For now - drill holes - it's only 15 min compared to years of knowing you did the whole amp yourself!
Now that I've etched PCB, I'm going to have a hard time buying commercial PCB.
Quick updates
Speakers:
I've finished the circle jig and layout for my speakers. I hope to have them up by mid-week.
Case:
My case front/back and top/bottom arrived. They still need some polishing, but I may have a case by Monday.
PCBs:
I'm going to etch a set of Balanced Cascoded F5 boards. Recall, I jury-rigged my dual channel Cascoded F5 into a single channel Balanced Cascoded F5 for testing. It's time to bite the bullet and do the full set.
Outputs:
I have a line on a set of 2SK2013/2SJ313s. Also, the 2SK3497/2SJ618s look deadly. I have to get a set first chance I have.
Grins:
I trying not to laugh, but sometimes it's too easy - I took a look at the "experts" Balanced F5 thread. It looks like they're going to use a Conrad MF35-151.5 based solution. The same heatsinks I've been using all along - funny how that worked out for me! They're thinking of adding a bipolar cascode to the board. I have a bipolar cascode on my Balanced F5 - funny how that worked out for me!
I guess I'm just really lucky all the stuff I've been doing for a couple of weeks is where the other guys are heading after months of rumination. It's better to be lucky than good.
Speakers:
I've finished the circle jig and layout for my speakers. I hope to have them up by mid-week.
Case:
My case front/back and top/bottom arrived. They still need some polishing, but I may have a case by Monday.
PCBs:
I'm going to etch a set of Balanced Cascoded F5 boards. Recall, I jury-rigged my dual channel Cascoded F5 into a single channel Balanced Cascoded F5 for testing. It's time to bite the bullet and do the full set.
Outputs:
I have a line on a set of 2SK2013/2SJ313s. Also, the 2SK3497/2SJ618s look deadly. I have to get a set first chance I have.
Grins:
I trying not to laugh, but sometimes it's too easy - I took a look at the "experts" Balanced F5 thread. It looks like they're going to use a Conrad MF35-151.5 based solution. The same heatsinks I've been using all along - funny how that worked out for me! They're thinking of adding a bipolar cascode to the board. I have a bipolar cascode on my Balanced F5 - funny how that worked out for me!
I guess I'm just really lucky all the stuff I've been doing for a couple of weeks is where the other guys are heading after months of rumination. It's better to be lucky than good.
Funny
Isn't that "Experts" thread yours? Funny The conrads have been talked about since this amp appeared, 2 1/2 yeas ago, along with the BVds of the input JFETs and cascoding the input? Didn't you read that? Sorry, if I'm making people laugh
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.