Huh? There is written:
If this regulator is used after CRC board, then only 100 uF is at it's output and regulator's impedance is between reservoir capacitance and amplifier as a dynamic load. If it is inserted instead of resistor in the CRC supply, then there is 60 – 80 mF at output and that is a completely different situation.
If regulator has low output impedance over audio bandwidth, 100 uF is enough.
Hi, your Gerber files does not have any outlines, so JCBPBC can not make the board.
Thanks
Dieter
Thanks
Dieter
Thats correct for the NPN
But, the 2N6287 PNP seems very expensive. When and where was the last time you bought this PNP for a similar price as the NPN?
Dieter there will no problem as I have ordered many pcbs by them but in any case I have send you the new gerbers with ouline included. check your pm.
Last time I checked mouser, there were no availability. I have sometime purchased it from the local market for 7 euros. It's a cheap choice.
That might have been true,
but thats not the current situation.
If there isn’t an alternative PNP this project is DOA for most potential builders.
but thats not the current situation.
If there isn’t an alternative PNP this project is DOA for most potential builders.
Vunce,
If I can find it at the local market of Athens - Greece, you will find plenty in US and the rest of EU. here is the 2N6286 which is identical to 2n6287
https://www.aliexpress.com/i/2255800967444980.html?gatewayAdapt=4itemAdapt
If I can find it at the local market of Athens - Greece, you will find plenty in US and the rest of EU. here is the 2N6286 which is identical to 2n6287
https://www.aliexpress.com/i/2255800967444980.html?gatewayAdapt=4itemAdapt
MPSA06/56, TIP41/41 (BD139/140), MJ15024/15025 (2 in parallel for each rail and big heatsinks).... no need for Darlingtons.
For positive rail only and exceptional performance... see my previous post (MPSA18, 2SD313 and 2 X MJ15024).... but 2N2219A (Philips) in place of 2SD313 will do amazing things... with a small caveat regarding maximum current drive - it also needs a little TO-39 heatsink. MPSA18 - there's simply nothing that can approach that thing - amazing. Real shame there's no complement.
The C1 should be max 47uF. Silmic II, or Black Gate N-type (bipolar). Each type will produce a fairly different perceived result.
The front end should be fast... CLC: 2200uF, L of your choice rated for 6A, second C also 2200uF. I used Black Gates FK types. I burn 10V across TO3 transistors, so things can get steamy when I run the rail at 6A....
The front end filters all the HF rubbish... the series regulator... well.... everything else.
No caps should be paralleled.
For positive rail only and exceptional performance... see my previous post (MPSA18, 2SD313 and 2 X MJ15024).... but 2N2219A (Philips) in place of 2SD313 will do amazing things... with a small caveat regarding maximum current drive - it also needs a little TO-39 heatsink. MPSA18 - there's simply nothing that can approach that thing - amazing. Real shame there's no complement.
The C1 should be max 47uF. Silmic II, or Black Gate N-type (bipolar). Each type will produce a fairly different perceived result.
The front end should be fast... CLC: 2200uF, L of your choice rated for 6A, second C also 2200uF. I used Black Gates FK types. I burn 10V across TO3 transistors, so things can get steamy when I run the rail at 6A....
The front end filters all the HF rubbish... the series regulator... well.... everything else.
No caps should be paralleled.
Last edited:
The ground returns from C1s should not join together the way you have them joined/tied together on the PCB... I will have a better look tomorrow.
Also, the in-ground from the front end and out-ground to the load are not shown.... it is important to keep those currents completely separate from that PCB. The ground tracks are way too narrow to be effective at HF... the output impedance/speed at high frequencies will suffer
Also, the in-ground from the front end and out-ground to the load are not shown.... it is important to keep those currents completely separate from that PCB. The ground tracks are way too narrow to be effective at HF... the output impedance/speed at high frequencies will suffer
I want to add that the elektor circuit used a 10000uF, which I have removed because of CRC, at the input.
I would like to declare ,once again. that this thread was created after maximax request.
Whoever thinks that this circuit can be improved by any means is more than welcome to go ahead. If you think that is a bad or unreliable circuit, you are kindly requested to focus on another thread.
Thank you
Whoever thinks that this circuit can be improved by any means is more than welcome to go ahead. If you think that is a bad or unreliable circuit, you are kindly requested to focus on another thread.
Thank you
Member
Joined 2009
Paid Member
Fyi - the pcb can be half that size if the power devices are off-board, just connect it with wires.
I am not sure why such a comment.... If you do not want suggestions, do not post here and ask for opinions.
The PNP TO3 you suggested is hard to find... I provided options for both rails - no Darlingtons. The transistors I suggested are still relatively easy to find.
I see that you used wider ground tracks for ground fill as I suggested, but the C1 placement is very wrong.... the same way the thin ground traces affect things.... well, the fuse links would do exactly the same.
My PS has TO3s mounted off the PCB, as Bigun suggested. There are multiple benefits to this approach...(only) one of which is -> you can double-up on TO3 very easily on an external hetasink.
Last edited:
If you want to use power darlington's - then you would be wise to redo the pcb based on the On Semi MJH6284G/6287G in the TO247 package. These are readily available and rated at 100v/20A/160W. So basically the same as the obsolete and hard to find TO3 types you have at the moment.
That is my thoughts on the pass device. Using these, you might find wider acceptance of the design.
If you want to stay with the TO3 types, then I would look at the available On Semi MJ11015G/11016G darlingtons.
That is my thoughts on the pass device. Using these, you might find wider acceptance of the design.
If you want to stay with the TO3 types, then I would look at the available On Semi MJ11015G/11016G darlingtons.
Last edited:
Hello,
I thought I was very clear from the beginning, that this was an old schematic, partly taken from Elektor mag. My 1st approach to the PCB was without TO3 on the board. After Maximax77's suggestion, I posted it here because it is (was) a low-cost cost reliable, dual rail PS for class-A amps. All opinions are welcome and if someone can introduce a better cheap circuit that could support class-A amps it would be a very interesting thread. This circuit works at my 4 class A amps and there is really no problem at all. When you suggest, that C1 placement is wrong and the thin ground lines affect normal operation you do have to explain the reasons. Don't you?
As far as the dimensions of the PCB, the gerber files are available to all of you, get them and create the PCB the way you like.