Hiraga "Le Monstre"

Jump to Latest
N

njepitt

Member
Joined 2008
Paid Member
danieljw said:


Sorry i was not clear, I meant in the shcematic the "Y" might be a typo. because the normal (as you have seen) groupings for the j74/k170 is GR, BL and V. per the datasheet.

i think perhaps the original article has a type and Y should be V???

I have not yet tried the 2SA1186/2SC2837 combo.

currently you are running if i am not mistaken:

2SK170BL, 2SJ74BL
2SC2240, 2SA970
BC640, BC639
2SC5200, 2SA1943

this is what i think after briefly looking at the data sheets.

2SK170BL, 2SJ74BL should function well.
2SC2240, 2SA970 relatively similar if hfe is selected appropriately
BC640, BC639 hfe is much to low (in my opinion)
2SC5200, 2SA1943 should function well.

-Dan
Click to expand...

I had understood differently; that really the "Y" grade, while available, was lower Idss than the others - only I can't remember (or find) where I saw this - can anyone clear this up for sure? If it's a typo, the same typo is repeated not only four times on the schematic, but again twice on the sketch of the pcb, which seems a little odd...

Actually I have switched 2SC5200/2SA1943 for MJW3281/1302, but other than that your list is correct. (I switched to eliminate the possibility of fake Toshibas... switiching back is relatively simple if indicated.) Any suggestion of easy-to-find drivers in place of B640/639? I have BC550/560 on hand, for instance, but the datasheets seem to give hfe for different input data, and I don't know how to convert...

Lumba Ogir said:
Hi,
Idss classification GR=2.5-6.5mA, BL=6-12mA, V=10-20mA.
Click to expand...

Lumba, I am not sure what you meant by the schematic you attached... Are you querying values of R1 and R2? and voltage across R3 and R4? Sorry if I'm being dense...

Cheers

Nigel
 
danieljw

danieljw

Member
Joined 2005
Paid Member
njepitt said:


I had understood differently; that really the "Y" grade, while available, was lower Idss than the others - only I can't remember (or find) where I saw this - can anyone clear this up for sure? If it's a typo, the same typo is repeated not only four times on the schematic, but again twice on the sketch of the pcb, which seems a little odd...
Click to expand...

I agree and would like to know this also.

njepitt said:

Actually I have switched 2SC5200/2SA1943 for MJW3281/1302, but other than that your list is correct. (I switched to eliminate the possibility of fake Toshibas... switiching back is relatively simple if indicated.) Any suggestion of easy-to-find drivers in place of B640/639? I have BC550/560 on hand, for instance, but the datasheets seem to give hfe for different input data, and I don't know how to convert...
Click to expand...

BC560B (specifically "B" suffix) looks like the right hfe range
however the Pd looks a little low 500mW vs 750mW, Cob is higher also
 
N

njepitt

Member
Joined 2008
Paid Member
Well it's too late tonight to try it out, but I can easily switch the BC550/560s in to see if they give an improvement... They don't seem to be graded, at least not visibly, so I'll have to check the Fairchild datasheet to see hfe and so forth.... Stay tuned...

Cheers

Nigel
 
N

njepitt

Member
Joined 2008
Paid Member
OK, here's an update. I followed Lumba's suggestion first. Measuring voltages as suggested, I had about 2mA going through the jfets, so I fiddled the resistance values on the trimpot to get things in order. Of course changing this changed the bias current also, so I had to alter those values.... et cetera. (Lumba: your schematic didn't show the bjts between R3 and R4 and the jfets - but these don't make any difference, right?)

In the end I got to about 1.15mA through the jfets, with bias current of about 550mA, by choosing R3 and R4 at 980R, and R1+R2 (total trimpot value) as 400R. (I expected this to be lower, but that's the way it worked out.....

Sound was helped a bit, but still louder than expected.

Then I decided to switch out the BC639/640 drivers. Turns out that the diferent pinning on the BC550/560 gave a (non-interesting) poroblem given my board layout, so I decided to try out 2SA970/2SC2240 here. (Looking at the datasheet the hfe was supposed to be about the same as BC550/560, and they are claimed to be "low-noise"...) This made a nice albeit subjective difference to the sound. A little less "stressed".

Still, much louder than the first Monstre. Let me repeat a question I asked above... What's an easy way to estimate an amp's output? Maybe the first one is weaker than it should be?

I should add that I don't think there is anything grossly wrong any more - with help from you guys the sound overall is pretty nice, now, although too loud. (It's also not as good as my F5, or as good as I had hoped. Well, maybe that's the way things go...)

Cheers

Nigel
 
N

njepitt

Member
Joined 2008
Paid Member
I've just been looking at a Le Monstre version at this site:

http://belinfabien.free.fr/amplificateur/le monstre/schema/schema_ampli.png

In this circuit he has resistors R14 and R15 on the bases of the driver transistors. What would be the effect of this? Am I right in thinking it would reduce voltage at base, hence reduce current through the driver, hence reduce gain in the output transistor? (Sorry, I realise my grip on these things is weak at best... )

Cheers

Nigel
 
N

njepitt

Member
Joined 2008
Paid Member
Lumba Ogir said:
Nigel,
if you are using BL devices, the source resistor values will necessarily be higher for a certain current. The biasing points are now right (videlicet as desired). R6/7 set the closed-loop gain, you could try a lower value for R6, say 180 Ohm instead of 220 Ohm.
Click to expand...

I think I understand this... reducing R6 means more feedback, which mains reduced gain, right? Sounds like an interesting thing to try.

This isn't the same issue I asked about, though - so you don't like the idea, or I misunderstood? And if I msunderstood, why does the schematic I posted have the 240R resistors on the drivers bases ?

Too late to change R6 values tonight.... It'll have to wait till tomorrow.

Cheers, and thanks

Nigel
 
L

Lumba Ogir

Banned
Joined 2005
Nigel,
answering your questions:
your schematic didn't show the bjts between R3 and R4 and the jfets - but these don't make any difference, right?
Click to expand...
Right.
I think I understand this... reducing R6 means more feedback, which mains reduced gain, right?
Click to expand...
Right. You may want to go even lower, towards 100 Ohm, using 1W resistors.
This isn't the same issue I asked about, though - so you don't like the idea, or I misunderstood? And if I msunderstood, why does the schematic I posted have the 240R resistors on the drivers bases ?
Click to expand...
There`s no need for base resistors here.
 
N

njepitt

Member
Joined 2008
Paid Member
Lumba, your suggestion certainly helped to cut gain down. In each channel I paralleled two 470R with the 220R already in the circuit, bringing the total value down to 112R or so. The amp is still a bit loud, but in a more practical range - by which I mean not so low on the volume control it's impossible to get right. Tomorrow (time permitting) I'll get a couple of 100R 1W resistors and do things properly. (At present they're just hacked on to the back of the boards...

Thanks

Nigel
 
stoolpigeon

stoolpigeon

Member
Joined 2007
Hi Dan,

In post #358 you seem to query if the Y grade of fet is correct but in the link (at the bottom of your posts) to the Jean Hiraga Le Monstre you talk about building this amp using the original fets which are shown in the schematic as the Y grade. Can you let me know which grade you used?

thanks, sp
 
danieljw

danieljw

Member
Joined 2005
Paid Member
the first one i built used y or V grade.

unsure if it is really a Y or a V though the writing on the jfets is so small...

recently i am unsure if it is y like the hiraga article suggests or a v like the data sheet suggests.

also i dont have the amp anymore so i cant check !!!

sorry for no help !!!

-Dan
 
You must log in or register to reply here.
Top Bottom