Aleph-X: High-Power Version

[Variac]

1. Is there a rule of thumb for calculating an effective value of the resistor in CRC Power supply filters? Is more better to the point- when your voltage gets too low or they heat up too much? Those issues I can calculate., but what is a reasonable value that is efective?


2. Now if you use a thermister, it has a range of course, I know Nelson has a mentioned the model he uses at some point. Can anyone help me track it down?

[nobody special]

I think that the CRC is basically an attenuator, so the more resistance you have, the better (keeping in mind what you said about heat and voltage loss).
I don't know if this will help you, but the datasheet gives resistances for different percentages of max current through each device:

CL inrush suppressors

[Variac]

I didn't search hard enough
here's some info:

Thermistor for Aleph 3 (Inrush current)

[macka]

Hi power X Aleph buddys,

Your opinions and comments are welcome & appreciated.

I am considering the power supply for my Aleph 2 conversion to the X Aleph.

I plan to use 2 300 va 24 volt transformers per mono block configured with duaL bridge rectifiers for 25 volt rails under load.

My X Aleph design is based on 25 volt rails and for maximum power 139 watts @ minimum speaker impedance of 7 ohms.

Therefore the Peak current is about 6.5 amps and peak voltage 43.75ac (derived from the 25 volt rails).

Each half of the amp will need to deliver 21.87 volts peak and 6.5 amsp peak, allowing 2-3 volt losses on the fets and source resisitors so the 25 volt rails are appropriate.

Given the 50/50 current share of the Aleph CCS, the power supply will need to deliver 6.5 amps x (2) @ 25 volts dc = 325 watts .

Here is the simulation on PS2. I have based this on the worst case senario (min voltage to the amp).

Using the existing capacitors 15,000 uf 63 volt x (4) per rail rated at about 20 amps each, I have opted for CLC filtering using 3Mh with 0.40 ohms DCR on the chokes (iron bar laminate sub woofer chokes). So I have 30,000 uf + 3Mh + 30,000.

The ripple is about 36millivots @ 6.5 amps.

Your comments an opinions most welcome.

macka

[macka]

With some careful re wiring I should be able to use the exisiting power fet boards. Hopefully the Fets will also be a reasonable match as I previously matched them at 0.50 amps. But that remains to be seen.

The X Aleph will of course be running at about 1.00 amp each Fet @ 25 volts (3+3 IRF 244 per side) with 0.5 volts across the 0.47 ohms source resistors. All I need to do is carefully cut the track on the power boards in the centre, Thankfully, there is provision for this in the layout!!

The two 300 va transformers will be stacked horizontally and the X board will replace the Aleph 2 board.

I'm sure it won't be that simple.

macka

[grataku]

Macka,
I know you don't care for MO but I think that your PS is not a good idea.
Specifically the DCR of the coil and the ~20ish watts it's going to dissipate in Cu losses alone. Plus if it's ironcore is probably going to saturate and generate even more heat. I guess you could split the core and have a mm or air but that will drop the inductance making the whole inductor idea pointless since what is doing the filtering work is the DCR not the XL.
Remember the dangers of having more iron in your amp too, more chance for EM noise.

[macka]

Grataku,

On the contrary,

You have raised some practical issues which I had not considered.
These fancy models are great for simulation but dont always help in the real world always!!

The heat is a problem you are quite right. I hope the nylon former does not melt or catch fire .

I will have heat with CRC also, so I figure the CLC at least would offers less hum.

However, as you quite rightly point out there could also be problems with EM noise. I recall someone having noise issue in a previous post...I don't recall who and Mr Pass referring to coil mechanic noise.

This is not good for a nice class A amp where quietness is the key.

I'm note sure about the saturation of this type of bar (straight iron laminate choke), they use then for high power subwoofers.

Over the next few days I will try both methods out to see what works out the best using the Baby X to start with.

I have plenty of spare voltage but would prefer not to go below 25 volt.

Maybe the Thermistor is the best after all!!

macka

[wuffwaff]

Macka,

I´m planing to use air chokes 1mH/0.12 ohms for up to 10A of total bias. These won´t saturate and the power loss isn´t that big. Using PSU2 this was the best I could get away with, without using too much space.

The setup will be 33mF-1mH-82mF


william

[grataku]

wuffwaff,
any particular reason to have the bigger cap after the choke? I would think it's better to have the lower impedance arrangment, ie the lower ESR where there is more ripple current to be handled, which is in the cap immediately following the rectifier. If my memory is correct the ripple current distribution is about 80% to 20% between the first and the second cap.

[nar]

Hi to everyone ,

I am getting started with my definitive version for my Aleph X .

Mono blocks version .
PSU +/- 24V , bias a little over 6 amps.Total of 12 IRFP240 per amp , with 0,5V or so on the 0,47 ohms source resistors .

I will set the AC current gain to 49 % .
I will use 3 heatsinks at 0,25 degrees per watt ,that on each mono block .

PSU will be 2 toroids ,2 bridges , one for each rail and total of 176.000 uF in 8 cans of 22.000 uF,arranged in C-R-C-R-C-R-C configuration .

But I wonder , on the XA 200 seems that Nelson uses zeners at the input preventing from electrostatic spikes , but I can see only 2 of them .Instead of 4 supposed to go between inputs and ground .

Is it possible to put only 2 of them, between gate and source of the 2 mos ?In that case is a 18 V model appropriate ? Does it protect the mos on both polarities of an eventual input spike ?

Thanks for answer

Regards