The Mini-A

[winslow]

I came across this interesting post a couple of days a go, and had one question- I want to build two of these for my car, and I have a couple 12 volt power supplies that do +/- 15 volts and 1 amp of current. Would something like that suffise, or would I need to find something else that has more current? I'm only needing 8 or so watts per amp to drive my compression drivers.

 

[mpmarino]

I came across this interesting post a couple of days a go, and had one question- I want to build two of these for my car, and I have a couple 12 volt power supplies that do +/- 15 volts and 1 amp of current. Would something like that suffise, or would I need to find something else that has more current? I'm only needing 8 or so watts per amp to drive my compression drivers.

Well, I'm no expert, but I have built a mini-a <and> I have done compression drivers in cars back in the early USD days. (hey, maybe my opinion will hold water!)

I think a mini-a would sound fantastic on compression drivers..as the old low powered soundstream class a stuff did. Those drivers were super efficient and sounded a little (or a lot) HOT, so a mini a may be a good choice. My only concern is the pSU you are proposing. The mini a as described will probably tax your supply more than you'de like. Others can probably comment on how it may sound with less bias or if it is practical to bias it back.

Marc

 

[winslow]

Thank you.

Yeah, I think it would sound awesome on the horns. I compete IASCA for Team Image Dynamics, and I'm always looking for that extra edge up. I can get the Image Dynamic power supplies, which can do either +/- 21 volts or +/-15 volts at .5A, however I can parallel those for more current- I had to do this when we converted my Rane RPM88 to work in the car. So, I could get 6 of those supplies and do 3 on each amp/channel and get 1.5A of current to them.

I also have a couple Monolithic A501M Class A monoblocks (which are a high biased A/B) and I could rob the supplies out of those since the are seperate from the amp board. These have +/-25 volt taps and enough current to drive the MiniAs, but I really wanted to keep output down (and heatsink size) for the car. I'm going to contact Monolithic and see if the have power supplies avaible seperately...or worse comes to worse, rob one from one of my amps to make a pair of MiniAs.

Anyways, I'm super excited about building these, and will find a way to pull it off b/c 10 watts of single ended class A on my 110 dB effecient horns would sound sweet. Heck, might build a couple boards and use some of Nelson's ideas from the F2 on one set of the amps.

 

[rmgvs]

question for mr Grollins (or Pass)

I am referring to your schematics diagram of the mini-Aleph of 12/28/01.

I have a question about the set of resistors and their values around Q4 and Q6. In your diagram you are using values of 750 Ohm for R 9, R10 and R12 in conjunction with values of .47 Ohm for the power resistors R18-R21.

In some other circuit of the mini-Alpeh I find values of 1k5 for R9-10-12 (but in conjunction with values of .33 Ohm for R18 and R19).

I understand that the value of R12 has to be seen in conjunction with the value of R20/21 (see articles about the Zen variations).

Now in a effort to build a mini-Aleph with the possibility to make it an aleph-3 some time in the future I have used the following values:

R9, R10: 1k5
R12: 750 Ohm
all power resistors .47 Ohm (4 of them, using only one pair of output transistors, 2 of these used in the output rail of the amp).

So in effect this is an Aleph-3 using only 2 output transistors in total.

Is there going something wrong here? What is used best in value for R9-10-12 using .47 ohm power resistors?

I am afraid of mixing up some circuits without knowing exactly what I am doing. By the way the amp works allright for a few days, so that is not the problem. Only want to know if I can do better by changing some values.

 

[BobEllis]

RMGVS,

I don't think changing R9 and R10 will have much impact on sound. But I may be wrong.

The ratio of R12 to R15 as well as R18 to R20/21 control the AC current gain. You may find a different value of R12 or R15 sounds better to you. You might want to measure the AC current gain and try different values.

 

[jleaman]

Here is a shot of one pair that i have been running on the new final psu..

 

[jleaman]

Here is a shot of the mini a's running nicly with my new dual dac and tube buffer..

 

[traw]

if only going for unbalanced in, is original schematic still what to go by except add 10pf across R8? seemed other cap at Q4 was debatable.

going back n forth between that schematic and briangt board schematic trying to settle on which one to etch n wire up.

also... if use 15vac x 2, 30vct... any resistors need adjusted?... it's 270va toroid not sure how much will sag. so somewhere between 18-20 volt rails.

 

[jleaman]

if only going for unbalanced in, is original schematic still what to go by except add 10pf across R8? seemed other cap at Q4 was debatable.

going back n forth between that schematic and briangt board schematic trying to settle on which one to etch n wire up.

also... if use 15vac x 2, 30vct... any resistors need adjusted?... it's 270va toroid not sure how much will sag. so somewhere between 18-20 volt rails.

U can use balanced or rca. Its a simple short neg input to gnd.. Maybe use a switch like i did.


Are you building mini a's or alephs ?

 

[traw]

i was thinking about giving the mini A a whirl... have aleph 30 built from briangt boards. have another set of the boards set aside for aleph 5. but was thinking about etching own board. seems can leave out more than handful components on diff input if don't need balanced input. so... just wanted to double check that first circuit posted complete, aside from adding the 10pf across r8 and also want to know if need adjust any values if using 18 volt rails (i saw grey's initial message said fine, just double checking)

 

[jleaman]

You don't want to go over 22v rails so that would be 22- 0 22+

I have 8 of these amps if you need any help just feel free to ask.

Jase

 

[MashBill]

I want to build an Aleph 30 with the BrianGT boards. How many IRFP240s should I order to end up with 12 that are VGS matched?

Or better yet, does anyone have VGS matched sets available?

Same goes for IRF9610. How many should I buy or does anyone have matched sets they want to part with?

I can PayPal the money to anyone that wants to sell me the matched sets.

Thanks,
Bill

 

[jleaman]

I want to build an Aleph 30 with the BrianGT boards. How many IRFP240s should I order to end up with 12 that are VGS matched?

Or better yet, does anyone have VGS matched sets available?

Same goes for IRF9610. How many should I buy or does anyone have matched sets they want to part with?

I can PayPal the money to anyone that wants to sell me the matched sets.

Thanks,
Bill

I might be interested in buying some with you as i'm building a pair of aleph 30's also with the briangt boards.

 

[GRollins]

I don't recall the Aleph 30 taking 12 matched devices. If I recall, each channel uses two banks of three. Each bank of three parallel devices should be matched, but you don't necessarily need to match the upper and lower groups to each other.
No one can guarantee a certain population of matched devices out of X unmatched devices, but you can probably do a pretty good job with a tube of 25.

Grey

 

[jleaman]

I don't recall the Aleph 30 taking 12 matched devices. If I recall, each channel uses two banks of three. Each bank of three parallel devices should be matched, but you don't necessarily need to match the upper and lower groups to each other.
No one can guarantee a certain population of matched devices out of X unmatched devices, but you can probably do a pretty good job with a tube of 25.

Grey

a Pair of aleph 30's takes 12

 

[carpenter]

So, it sounds like a stereo pair requires 4 matched sets of 3 FETs.

 

[jleaman]

So, it sounds like a stereo pair requires 4 matched sets of 3 FETs.

yup

 

[GRollins]

a Pair of aleph 30's takes 12

It takes twelve devices, yes, but all twelve do not have to be matched as a single set. Four sets of three will do very nicely. The number of MOSFETs required to get a full set of twelve is much, much larger than the number required to get four sets of three.
Incidentally, I would not count on being able to build four channels of amps out of twenty-five devices. Two channels is a fairly good bet, with some left over for Mini-As or voltage regulators or something.
Of course, all this depends on the luck of the draw; how tightly grouped the parts in the tube are. And how close you want your Vgs numbers to be.

Grey

 

[jleaman]

It takes twelve devices, yes, but all twelve do not have to be matched as a single set. Four sets of three will do very nicely. The number of MOSFETs required to get a full set of twelve is much, much larger than the number required to get four sets of three.
Incidentally, I would not count on being able to build four channels of amps out of twenty-five devices. Two channels is a fairly good bet, with some left over for Mini-As or voltage regulators or something.
Of course, all this depends on the luck of the draw; how tightly grouped the parts in the tube are. And how close you want your Vgs numbers to be.

Grey

True.. I hope to build a A30 to compare it with the 4 Aleph Mini's I have.

Chassis being ordered 2 of these

An externally hosted image should be here but it was not working when we last tested it.

 

[MashBill]

True.. I hope to build a A30 to compare it with the 4 Aleph Mini's I have.

Chassis being ordered 2 of these

An externally hosted image should be here but it was not working when we last tested it.

Nice chassis. Who sells them?