Please help with my F5/F5T questions

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I know very little about audio power amplifier design and construction. Most of my experience with audio electronics has been with line-level circuits, and although I am handy with a soldering iron and have built from schematics without disastrous results, with my level of ignorance and inexperience I won't build a power amplifier without more guidance.

As part of a general renewal of my home audio system I would like to construct a First Watt F5T variant, and I have a few questions that I need answered before I proceed. I have read both the original F5 article and the F5T article as well as much of the various F5/F5T threads, but I have not seen answers to my specific questions. I have tried to work out the answers on my own, but my mind begins running in circles from informational overload and my head starts to hurt! I hope to short circuit the process and find answers in this thread. I ask for your patience and indulgence if these questions have already been answered elsewhere.

My idea for the amplifier is to build two monoblock modules (each with its own separate power supply) to drive some DIY speakers. I have determined that for these speakers (and with the preamp I plan to use) the 22dB gain of the F5T V1 would be just right. In addition to the gain requirement of 22dB, the amp would need to produce about 11 volts and about 50 watts average (all in class A, I hope, as this would be the main reason for using the F5T). The speakers will have a reliable minimum impedance of 4 ohms.

These are my questions:

1. Can the F5T V1 be built in a point-to-point form, forgoing a pcb? I seem to remember reading in one or more threads that some DIY members have done so, but with the high bandwidth of the amplifier, would point-to-point construction create any problems?

2. I am a bit concerned about the dissipation of the input Jfets with the F5T V1 power supply of 32 volts. Is it possible to obtain the output requirements I have listed with the standard F5 power supply voltage of 24V?

3. Relating to question 2, how high would I have to bias the output devices to get to my output goals on 24V?

4. Does anyone have long-term experience with a 32V F5T V1 equipped with the original Toshiba Jfets? Did you follow the suggestions of Nelson Pass to preserve their service life, or did you do something different? Have you had to replace any of the input devices?

One of my main goals is keeping the circuit as simple as possible. This is why I would prefer to avoid cascoding the input devices, for example. That being said, if any DIYers have any ideas on how I can build an F5T V1 with the output capabilities I listed, I am open to suggestions. Thanks to all for any help you can offer.
 
1) yes it can be done. but you also can run into trouble if it's not done right.
2) you will acctualy be better of runing it on +/-24V rails. that will give you around 50W average at 4ohm.
3) to get 50W average class A at 4ohm, you need 2.5A bias.
with 2 output pairs you will have around 30W dissipation pr output fet.
 
Thanks, AudioSan. Could you or anyone else expand upon the point-to-point wiring concerns as related specifically to the F5T? I assume such things as "keep the inputs away from the outputs", but are there other specifics to consider? Would it be a good idea to limit the bandwidth, and if so, to what value?

Can I assume the figure of 2.5 amps bias is total bias, so that 1.25 amps goes to each single device in an output pair?

Does anyone have any other comments or suggestions?
 
An F5 has a single output pair.
The bias current passes from +ve supply rail through the two output devices.
The current is the same current in the top device and the bottom device.

If the total bias current for that single pair output is 2.5A then that same currents passes through the top device and then that same current passes through the bottom device.

Yoy DO NOT add the currents in the the top and bottom devices together.

If you have an F5t with two pair output stage.
The bias current in one PAIR and the bias current in the other PAIR should be added together to get the total bias current.
This total bias current flows from the +ve supply to the -ve supply.
 
Last edited:
  • Like
Reactions: 1 user

6L6

Moderator
Joined 2010
Paid Member
With your self-proclaimed inexperience I strongly suggest building your amp using a circuit board. There are many availaibale, the one from the diyAudio store and Peter Daniel's to name two, they are know to work and be stable.

To answer your questions specifically -

1) Yes you can. I my opinion I don't think it's a great idea since there are PCB available.
2) It's not really that much of a problem, but if you are concerned, that is exactly what the cascode circuit if for - protecting the Jfet from large rail voltage.
3) As audiosan said.
4) Keeping the inside of the chassis cool is going to keep the jfet healthy more than anything else. Lots of vent holes to and bottom.
 
AndrewT and 6L6, thank you for the additional information.

I want to be sure that I understand the bias process correctly. I quote from Nelson Pass in the F5 Turbo article: "The (output) devices are biased at about 1.3 amps in the original circuit for a dissipation of about 30 watts per device", and "...the F5 Turbo V1: Here you see the F5 without the limiters and with an additional set of output devices. With adequate heat sinking, you can bias them to 1 amp each, for a total of 2 amps, which will operate the circuit Class A well above 100 watts peak into 8 Ohms." This is with 32 volt power supplies.

Again quoting Nelson Pass, from the original F5 article: "For this sort of circuit, a 1.3 amp bias means that the amplifier will operate
Class A to 2.6 amps of output current......The power of 2.6 amps into 8 ohms is I^2 * R, or 2.6 * 2.6 * 8 = 54 watts. This is the peak value, and the nature of an undistorted sine wave is that the peak wattage is twice the average, so this circuit would operate 27 watts average Class A into 8 ohms." This is with the original F5 24 volt power supplies.

The F5T V1 has a parallel pair of output devices on both the positive and negative legs of the output. If I have 24V power supplies as in the original F5, and I have parallel output devices as on the F5T V1, and I bias each parallel pair to 2.6 amps (1.3 amps per device in a parallel pair), would I achieve an output of 54 watts average into a 4 ohm speaker load?
 
In the original F5 article Nelson Pass states the maximum unclipped output of the amplifier as +/-20 volts (this is with the 24V power supply). If I build the F5T V1 (with 24 volt supplies), they would be monoblocks, each with its own individual supply. Am I correct in assuming that form would give me the output capability I want?
 
6L6, I'm having difficulty understanding your reply, "Assuming you can swing 40V pk-pk, yes." Could you elaborate, please?

Perhaps it would help you to help me if I gave some more information. The DIY speakers with which I plan to use the F5T would need only 11.3V RMS to produce the sound level I wish to achieve, but due to power dissipation in the crossover, they would need almost 50 watts to do it. It is the Class A current capabilities of the F5T that make it attractive to me in this application.
 

6L6

Moderator
Joined 2010
Paid Member
If the amp is biased properly, I.E., the current that has been discussed, the power limit will likely be limited by voltage swing... 40V pk-pk is a few volts under the rails, which is a nice comfortable margin, and should give you the power you want.

40V Pk-Pk (another way of saying +20Vto 0 to-20V)


Power in watts is V rms squared / load --


40*.3535 = 14.14V Rms (.3535 is the conversion from pk-pk to RMS)

14.14^2 = 199.9

199/4ohm = 49.98Watts

In all likelihood, the amp will give a bit more power than that.
 
Last edited:
If I'm reading the schematic of the F5T V1 correctly, it appears that I could run the power supply rails at 24V without any changes to the circuitry. Could someone confirm that for me?

I would like to provide some circuit protection. I am considering using fuses on the outputs, just like they used to do back in the good old days. Are there any drawbacks to this?

I wish to again thank all those who have responded, as well as any who may yet do so. You are helping me to fulfill one of my "bucket list" aspirations.
 

6L6

Moderator
Joined 2010
Paid Member
Yes, you could run the F5T with 24V rails. No changes would need to be made.

Yes, you could fuse the outputs, no drawbacks other than the additional parts to buy.

Although you could kill a few birds with one stone and look to see if anybody has a cviller F5C PCB that they are not going to use - it holds 2 pair of outputs, can be configured to cascode or not, and still has the limiter circuit...
 
Last edited:
Woody,

Although the maximum power output may increase with lower impedance (many class AB amps maximum power nearly doubles with half impedance, but not always), the class A output is defined by the standing current (bias). So as Andrew says, into 4R the class A output is lower at a given bias. If you want more class A power you have to increase bias. At some point the amp will become voltage limited and then you will need to increase rail voltage and bias to increase the class A output. Higher rails will help max power at any load impedance, but you need both bias and voltage to keep it in class A.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.