There is a complete thread here on these amps. I think there were some suggestions on transistors.
Pass uses available transistors. That is the advantage of HEXFETS, you can get them. Laterals are only Exicon and Alfet. As there is nothing cheap about class A, maybe the cost of ordering FETS from England is not that much in overall cost. Part of the fun of DIY is adapting to what is available.
You should still look at the "Square-Law" concept in Linear Audio, Vol 1. I suspect there have been some threads on it.
You can take most any conventional class B amp with a decent driver stage and up the bias to class A. Reduce the rails to a suitable voltage. Nothing magic about it. Massive power supply, massive cooling. Of course witht he low voltage rails, you can do things like a op-amp input stage without difficulty.
I have a JHL driving 91db speakers in a medium room. More than sufficient for listening.
I up the efficiency and power by 6db (each) for parties. But then no-one is listening to the music and there's just a little more absorbtion in the room.
I considered just using some class AB amp and up the bias to class A. However, I don't know if most designs work well with just +-16V. For example the class A amps in Rod Elliott's page use higher supply and more power.
Anyways as for bias current and power consumption, I = U / R = 16V / 8 ohm = 2A. This would mean that quiescent current would be 2A. So, power consumption per channel would be 2 * 16V * 2A = 64W. So, the entire amp would use 128W power, so theoretically the 180VA transformer should be fine..
Anyways, which thread were you talking about?
Last edited:
hits
Cool, they still make the 2sk1056/j162. Nothing ever came up in a search but people in forums looking for replacements being told no such luck. Only big package, but that's OK. Now to find out where to buy them.
They just list the numbers. Nothing about their technology in the data sheet. Sloppy WEB site. Again.
Cool, they still make the 2sk1056/j162. Nothing ever came up in a search but people in forums looking for replacements being told no such luck. Only big package, but that's OK. Now to find out where to buy them.
They just list the numbers. Nothing about their technology in the data sheet. Sloppy WEB site. Again.
If you are going to spend a lot of effort on building a class A amp, I'd sort you transformer out first, rather than try to design something around the very low rails you have. For 15 W true class A, I am using 18-0-18, which after rectification, smoothing and unde a load of 1.4 A per channel gives me +-22VDC. I am using straight EF output stage.
My transformer is rated at 250 VA.
The JLH is a very highly regarded amp, as are Nelson Pass's designs. With the Pass designs, you have a lot of support if you run into any problems.
My transformer is rated at 250 VA.
The JLH is a very highly regarded amp, as are Nelson Pass's designs. With the Pass designs, you have a lot of support if you run into any problems.
You don't want to use a transformer at its rated capacity or that close. The internal losses will cause it to heat up rather high. This heat will reduce the life of the rest of the amp, smoothing caps in particular.
A rule of thumb is use a transformer at a bare minimum of twice your expected load. Three times is better. So for your 128W I'd go for at least 300 VA. I'm using 400 VA per channel on my Aleph-J drawing around 100W. Transformers are fairly inexpensive, pick the amp then buy a transformer to suit. Your transformer may be suitable for a single channel.
Hello. I wanted to share my experience with this amazing amp.
I made this amp in its original 1969 version there two years. The output capacitor is very affecting sound quality and this is its biggest flaw.
I did a lot of testing and the best result is especially not put a big capacitor (3300 uF) but rather several capacitors in parallel by gradually decreasing their value. In my amp I put 2x Philips 1000μf ll hp + 1 x 470 uf elna Silmic + 1x 220uF Panasonic FC + 1 x 47 uF elna Silmic +1*10 uf elna Silmic + 1 x 4 uF capacitor paper mbgo2 russian + 1x 1.5 uf capacitor polypropylen ERO and finish with a russian pio 0.33 uf k40y2. The input capacitor is also a Russian pio k40y2 1 uF (as good or better than jensen pio). It sounds clear and transparent and this simple circuit is still an advantage compared to the 2 power rails version, I made the comparison.
@+
I made this amp in its original 1969 version there two years. The output capacitor is very affecting sound quality and this is its biggest flaw.
I did a lot of testing and the best result is especially not put a big capacitor (3300 uF) but rather several capacitors in parallel by gradually decreasing their value. In my amp I put 2x Philips 1000μf ll hp + 1 x 470 uf elna Silmic + 1x 220uF Panasonic FC + 1 x 47 uF elna Silmic +1*10 uf elna Silmic + 1 x 4 uF capacitor paper mbgo2 russian + 1x 1.5 uf capacitor polypropylen ERO and finish with a russian pio 0.33 uf k40y2. The input capacitor is also a Russian pio k40y2 1 uF (as good or better than jensen pio). It sounds clear and transparent and this simple circuit is still an advantage compared to the 2 power rails version, I made the comparison.
@+
Last edited:
All true, especially with "consumer" transformers.
However there are transformers out there which are designed to run at their rated capacity day in, day out.
It's a bit like car engines vs aero/marine/stationary engines. Your car might be able to produce 500hp - but don't try to run it at that level for very long!
On the other hand the aero engine will run at 80% rated power continuously. And love it
But those are actually derated transformers.
I use four 1600VA (ex-military) floating isolation transformers for my audio gear, each weighs 55lb, the toroidal core is 3kVA minimum.
http://www.diyaudio.com/forums/atta...87426-pass-monster-separation-1600va.big4.jpg
Same same for commercial shore-to-ship transformers, used in yachts.
Those can be loaded to 100% of their nominal capacity, but weigh a ton each.
And take a look a Plitron's tech notes. They suggest that the current in the secondary of a transformer driving a rectifier and capacitor input filter will be 1.8 times the DC current. So the rule of thumb of twice actual draw has some merit. Then look further down the page and see how much cooler the transformer will be operated even slightly below its maximum rating.
Also look at the ESP audio site. There are three class A amps listed. The first is his version of the Zen. I tried it but was not too impressed.
Next is the 20w version. This is a true class A amp and not an over biased class AB. I built three of them for my LCR tweeters. They want +/-22-25v but should work OK at +/-16v with less power output.
The third is his famous project 3A but biased into class A. This does have an advantage that when pushed beyond class A, it will enter class B and keep going. Plus you can easily make this into a 60w class AB amp just by adjusting the bias. But it is not a true class A amp. It also wants +/-33v on average.
I'm an EE myself and find the ESP site one of the best online for good solid audio designs and information.
Next is the 20w version. This is a true class A amp and not an over biased class AB. I built three of them for my LCR tweeters. They want +/-22-25v but should work OK at +/-16v with less power output.
The third is his famous project 3A but biased into class A. This does have an advantage that when pushed beyond class A, it will enter class B and keep going. Plus you can easily make this into a 60w class AB amp just by adjusting the bias. But it is not a true class A amp. It also wants +/-33v on average.
I'm an EE myself and find the ESP site one of the best online for good solid audio designs and information.
the current will be much much lower.
you will have som voltage losses. so you end up with about 12V/8ohm=1.5A PEAK. quiscent current demand is half. lets see if that is correct.
12*0.707=8.5Vrms
8.5*8.5/8ohm=9Wrms (voltage limit)
0.75A bias = 1.5A peak.
1.5*1.5*8/2= 9Wrms (class A current limit)
spot on.
so. consumption will be: 16V*0.75A*2=24W pr ch.
so you can use your transformers without any problem.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.