The post2 schematic shows 2pair. That's where I got the information.
Is the post2 schematic wrong?
post 2 wrong?
Andrew,
I don't know that the post 2 scheme is wrong... I believe it has just evolved
Andrew,
I don't know that the post 2 scheme is wrong... I believe it has just evolved
Yes this helps. My tentative on paper build consists of two 400va 45v x2 antek transformers with covers. The antek covers take care of the shorted turn issue, though in my opinion only if mounted to a metal base of appropriate thickness.
Should this discussion be moved to the build thread?
i make my own traffos for all my amplifier builds.....i can make them any size i want.....
voltage sag happens the instant your circuit draws current, this is physics and this is unavoidable...
i asked the question in this forum many years ago about the proper sizing of power transformers, and the answer i got from the gurus, "get the biggest size that will fit in your box".....
there are other answers to that question in these forums......
the relationship between power transformer size and output stage robustness is a balancing act.....
get a small transformer and a robust output stage and your traffo burns out, get the biggest traffo and your output stage burns out....
now which is easier? replacing traffos or replacing output trannies?
the choice is yours.......😀
Last edited:
XFMR size
Tony,
There are many problems with the "get the biggest XFMR that will fit in your chassis" philosophy.
1. We have foregone doing a few calcs to come up with the correct size based on the application
2. The "law of diminishing returns" takes effect. Once the optimum XFMR based on calcs is put aside to opt for "bigger" the cost starts to become exorbitant for the amount of performance gained.
If you have a 20^2 Meter box, that XFMR would be pretty big...j/k
Tony, kudos for being able to wind your own toroids! I am sure that is quite an art!
Regards,
Brannon
Tony,
There are many problems with the "get the biggest XFMR that will fit in your chassis" philosophy.
1. We have foregone doing a few calcs to come up with the correct size based on the application
2. The "law of diminishing returns" takes effect. Once the optimum XFMR based on calcs is put aside to opt for "bigger" the cost starts to become exorbitant for the amount of performance gained.
If you have a 20^2 Meter box, that XFMR would be pretty big...j/k
Tony, kudos for being able to wind your own toroids! I am sure that is quite an art!
Regards,
Brannon
balancing act
Tony,
It does require a few calcs, but it's fairly straight forward to find the optimum XFMR size:
1. Power output
2. Power Losses
3. Power of auxiliary circuits
Output transistor's have SOA's, and heatsinks should be implemented that are representative of that. This is not as straightforward as optimum transformer VA, but it is achievable. There are several good texts (and posts on this site) that relate the laws of thermal dynamics wr to audio amplifers ( Self, Cordell, etc..)
Regards,
Brannon
Tony,
It does require a few calcs, but it's fairly straight forward to find the optimum XFMR size:
1. Power output
2. Power Losses
3. Power of auxiliary circuits
Output transistor's have SOA's, and heatsinks should be implemented that are representative of that. This is not as straightforward as optimum transformer VA, but it is achievable. There are several good texts (and posts on this site) that relate the laws of thermal dynamics wr to audio amplifers ( Self, Cordell, etc..)
Regards,
Brannon
re:
I would regard any amplifier that "burns up" transistors and transformers on a regular basis as an "ill conceived design", I'm sure you would too 😉
I would regard any amplifier that "burns up" transistors and transformers on a regular basis as an "ill conceived design", I'm sure you would too 😉
oh, yes.....but you know how these amps can be abused.....no design is ever bullet-proof in the hands of the abusers.....
years of troubleshooting and repairing commercial amps tells me so...😀
not meaning to be rude but......much as i would love to carry on this conversation, i am afraid there is nothing you can tell me that i did not already know......😉
The HB prototype currently drives a pair of 6 ohms Wharfdale 9.1 speakers, but it also drove a pair of 4 ohms Dynaudio Audience for about 2 days too. Aside from that, it has driven other bookshelves and floorstanders of different brands.
The secondary of the toroidal I used in the prototype is 45VAC.
So far the amp is still alive.
Is higher temperatures produced by the output devices at lower impedance loads the concern?
I am using a prototype of the soon-to-be-available DA 5U chassis. 8 or 4 ohms the highest temp I can read from my infrared thermometer is just 45*C. It was summer time when I took it and ambient can be as high as 32*C.
The highest temp I registered 49*C was when I biased the amp to 250mA per device.
Heatsink your amp well and it will be fine.
Different Rail Voltages
If I were to build two of these amps at different rail voltages 50V and 70V what components do I need to change (apart from the voltage ratings of the caps)?
If R14 and R36 need to be changed - what values?
Thanks.
If I were to build two of these amps at different rail voltages 50V and 70V what components do I need to change (apart from the voltage ratings of the caps)?
If R14 and R36 need to be changed - what values?
Thanks.
+/-70Vdc rails is too high for use into 4ohm speakers.
Because the frontend and VAS have CCS's, I wouldn't think any resistor values would need to change. The bias for the drivers will shift, but not substantially.
Because the frontend and VAS have CCS's, I wouldn't think any resistor values would need to change. The bias for the drivers will shift, but not substantially.
silly question; could additional output devices be used to support 2 or 4 ohm speakers at these types of voltages?
Thanks
Rob
Thanks
Rob
What current would be demanded of the VAS, if high voltages are used to drive low impedance speakers?
Right, this is a basic dual emitter follower design. Lower impedance speakers will put a higher load on the VAS than a triple EF output stage would. Simply adding more output transistors would help, but even at +/-70Vdc the VAS will be pushed harder than intended.
The amp was designed around +/-60Vdc rails. Unless you can do the calculations, and modify the design properly yourself, you should build it as the designer intended.