F5 Turbo Builders Thread

I reread the article last night. I don't get alot of details from it for consumers of this thread to follow. He is informing us how to do it but, in a way we can go figure it out for ourselves. That would include experimentation...
I started way back when cviller was doing his dual output board. I'm just slow if no one else is paying or wanting. That was right up my alley though. And, there are some of the important details in the Low Z thread not mentioned here. I want to paralell my line array drivers as much as possible. All 12 in paralell would be 1.33 Ohm nom. and that would be good if the amp can do well. I don't know. Conservativly though, 6 strings of 2 - 8 ohm nominal spk's makes 2.66 ohms. Total DCR of 1.86 ohms. I want to see how well I can do with that. Re wiring the spk's is easy fo further experimentaion. 240w of dissapation from my stereo f5t Ver 1 amp will only add to the 180W dissapation of my tweeter amp. I have an 8 ohm center channel woofer too.
My problem with what's going on in this thread is, it's not what I want or need. I just wonder if it is actually what most of us need? And, of coarse what most of us want I'm not going to address???
The real problem is, no one can specify an operating point that will be good for everyone. Taking the Bear voltage test on your speakers will get you started but, there is just to many variables with efficiency, room, music style, avg volume, etc. to pin down a transfer curve that is ideal for everyone.
First I would start with an F5, or maybe a dual, or tripple. 4 outputs has to be about it. I would then realize (in my case) I'm using 2SK/J parts that are only 12A Max Id/FET. As is the P channel Fairchild and??? So, I'm comfortable with 1.5-2A Iq. With the dual output cicuit, thats 6-8A ClassA out. Maybe the continuous peaks during a few friends over a few bears, or a barbeque in the yard, we hit class AB at least every second up to maybe 12-16A peak. No problem. I now have a a small margin for the 1812 oveuture or one of them visiting barbie freqs to turn the wrong knob, play the wrong source, or the wrong tune or whatever, before the Outputs are more done than the steaks!. I don't need no stinking diodes!
 
The diodes would seem to fit better with less output pairs vs more. Cascoded V2 seems teh most practical to me for high power, but some folks have 3 and 4 way speakers, with multiple hungry woofers in them. Add to that crossover losses, and maybe you need a V3. Beyond that ...?
If some my parts come in, I will have fully operational, 2 pair V3. Finishing the cab up now. I will put scope on diodes and play music and see what happens. May have to short the outputs to get the diodes to conduct:D. I guess me and the kids could have funny hair day with Dad's amps. Her son, touch tese two together.
 
Will someone double check my PSU. I am calling it Lerch
 

Attachments

  • PSU.jpg
    PSU.jpg
    346.9 KB · Views: 902
After ZM's mods, rotate the left two diodes 90 degrees clockwise, rotate the right two diodes 90 degrees anticlockwise, plus swap the position of the two diodes on the right side of the board to get rid of the links, then move the transformer connections to a vertical line.
 
you can bet that drivers in your arrays are soooooooooo stupid , that they're simply no aware are they connected in series or parallel

:rofl:
Are you saying that 2 or many drivers in series will not have effects from sharing the same box? Or that electrically, they do not suffer ill effects from the others being in series with them? Or are you suggesting those effects I spoke about, are not present at all? :D
 
Thanks guys. Clearly a newb. Why is it better to have (+) and (-) in central area of pad over edges as shown. Also, why is it better to have a ll grounds at central point as long as they are on same pad? Last;y, if going double sided, do you want ground plane directly under positive and negative planes

Audio San,
Its not that the thermostats have not been in amps, just that I (NEWB!) have not seen them.
 
Will someone double check my PSU. I am calling it Lerch

Almost good
Stick the diodes on the vertical side instedad of botom

Line up the caps and it is done bigh thick and shoorth cooper traks is what is needed here(exactley as you did)

I am using different diode bridges so did not bother with finishing that part off.

I will be Happy to do it but you doing prety good job so no problem there...

Diodes on left side output on right red outline for 5 resistors and final smmoting film cap.

There is plenty of clearence for larger size caps
 

Attachments

  • CAP BOARDS 001.jpg
    CAP BOARDS 001.jpg
    665.4 KB · Views: 828
Last edited:
If you are tring for single sided which is IMO better there is no going around one of the track on the diodes crossing over the other.
I spent few hours triing to figure out a way which alwaise turn up to be a Jumper wire.

If you go duble sided I would Mirror the tracks on the botom layer (apart for where jumper wire is ini't) current carring capacity is the point here no screaning

If you want screaning from nasties DC filter and chokes is way to go
 
Official Court Jester
Joined 2003
Paid Member
Are you saying that 2 or many drivers in series will not have effects from sharing the same box? Or that electrically, they do not suffer ill effects from the others being in series with them? Or are you suggesting those effects I spoke about, are not present at all? :D

series or parallel , they don't care and they don't know

but amplifier knows , same as effect on cable gauge and critical length
 
Thanks guys. Clearly a newb. Why is it better to have (+) and (-) in central area of pad over edges as shown. Also, why is it better to have a ll grounds at central point as long as they are on same pad? Last;y, if going double sided, do you want ground plane directly under positive and negative planes

Audio San,
Its not that the thermostats have not been in amps, just that I (NEWB!) have not seen them.

aha. they are ginuiss in class A:) if you have a run away, they cut your power and save your amp and/or speakers:)
but alot of people(me inkluded) are not a fan of having AC wires run all a cross the amp in every direction:) so a smal signal one is the way to go:)
 
The diodes would seem to fit better with less output pairs vs more.
I think the trend would be in that direction also. I think it's important to start with something as close to F5 as possible. Remember the 1st watt? I understand wanting maybe 50-100Wrms classA. That's the area of what is easy enough to work out or WAF or whatever. You just have trouble dissapating the necessary heat that comes along with an amp any bigger, that is classA.
So, if you have a need for more power, say 10 times the power of Nelson, trying too build 10 of Nelsons amps to get 10 times the power will need at least 1500Wrms from your wall socket! Actually, the F5 is a fairly efficient design.
IMHO you can build a very good amp without much more complexity than the original that does 50-100Wrms in classA and 2-300Wrms by sliding into classB and shunt diodes. If your power supply and Voltage rails are configured accordingly. And your circuit Iq and FET Idmax are suitable. In other words, if the amp is as close as it can be to the F5 to get you your 50-100Wrms into your load, the amp will be stable at it's nominal design point but have the ability to momentarily output clean peaks 2-4 times what the classA parameters would support, without 2-4 times the heatsink. You are gonna need more heat sink. ClassB produces more heat with more output power unlike ClassA which does not get hotter when the music is playing. You also won't have double or tripple the electric bill from your air conditioner cooling the room these amps are heating up. Find a comfortable quality operating point that does not dissapate more than you can tolerate. I have had a 750W heater in my listening room. I only need to turn it up 1/2 way in the winter to heat the room!
:D