| AudioFreak |
| Use power dissipated by a single FET. |
|
|
| JDeV |
| quote: | Originally posted by Algar_emi
My problem, if a problem it is, is that when first starting the amp, The current on the Positive side of the supply jumps to the maximum of my test supply (limited to 3A for testing), then after about 1 sec, it drops to the normal 1.5A.
Is this normal?
If not what can cause this behaviour? |
Do not worry, very normal. I just finished my A30 as well, 2 days ago - same problem. The current jump is due to Inrush current, which is from the transformer. (I measured this inrush current with a peak crest meter and it was as much as +33A and -17A on the 1st few cycles. This is for onle a split second then it starts dropping.) Best solution are to fit a slowblow fuse AND add a thermistor inline just after the primaryside fuse. (As in Aleph3) This method works perfect. More about this here:
http://www.meritekusa.com/meritek_p...rThermistor.pdf
http://www.kristronic.com/softstart/tech.htm
http://www.tortech.com.au/article.html
Post some pic's if you got and good luck with your project. |
|
|
| Algar_emi |
Thanks for your very usefull answers.
Bye. |
|
|
| JDeV |
| quote: | Originally posted by UrSv
Yes, lowering R19 will decrease the bias. Try connecting a resistor in parallell with R19, say 470K, and see what happens (you can do that with alligator clips for easy removal). Remember that you may have to adjust R21 for the current source gain gain.
/UrSv |
Swede: Did you actually change R21 on your amps and what was the result? |
|
|
| swede |
Hi,
I have not yet have had the time to fiddle around with R21. Though I've changed R19 on one channel, to get it to dissipate as much heat/biasing it as the other one.
Though I will have a look on R19 some time after christmas, just to get the grip on it.
//magnus |
|
|
| JDeV |
| quote: | Originally posted by swede
Though I will have a look on R19 some time after christmas, just to get the grip on it.
//magnus |
You mean R21 ???
If I may quote Peter Daniel:
"Say what you mean-mean what you think-think anything
Why not"
;) |
|
|
| JDeV |
Why do I see this where Swede's signature is suppose to be?
"__________________
perl -ne'chomp;$a.=pack"b*",$_;END{print"\n$a\n\n"}'<<RM
10000010001101101010011000001110000101100000010001001110
101011100011011010100110110011101000010001010000
RM"
Just out of curiosity. |
|
|
| UrSv |
It IS his signature...
/UrSv |
|
|
| JDeV |
| A-ha! So he got me :) |
|
|
| swede |
Run the perl command in any UNIX like terminal window. ;=) It is not harmful.
//magnus |
|
|
| gnomus |
Just got my aleph 30 going. I am using it single ended. There is hum that I can just hear from the listen position about 10 feet(3 meters) away. My speakers are not supper efficent. So this seems a bit much. Is this normal or to much hum? My power supply is an exact copy of the original aleph 30. My other impression is that this is a pretty dark sounding amp. Pleasant sounding and smooth but covering up detail.
Thanks in advance. |
|
|
| Algar_emi |
Here a small table listings the value that I found in my Aleph30:
R19 I-FET PWR
47.5K 0.8A 20.0W
33.2K 0.6A 15.0W
30.2K 0.5A 12.5W
26.7K 0.4A 10.0W
Power dissipation is for 25V Rail...
Bye.
SB |
|
|
| wuffwaff |
Paul,
hum should be inaudible at more than a few cm from your speakers so if you hear it at 3 meters something is not quite right.
Dark sounding? I can imagine what you mean but can´t really find this in my Aleph5 (wich should sound more or less the same). Of course it depends on where you´re coming from (amp wise) but I must say that one of the things the Aleph does really good is revealing small details in the music that go by unnoticed with lesser amps. It does this without sounding aggressive or bright.
How have you set your bias and ac-current gain?
william |
|
|
| Koy |
| Gnomus: Thanks. I have the same feeling of the loss of details. I believed ( according to furore of other diyers and rewievs) it could be something wrong on my side. In all probabillity not. Thank you for you oppinion once more. By the way it seems to me that the older Aleph 3 sound better than A30 in this point of view. So I rebuild A30 to A3, rised bias (to 0.9A per case and I am experimenting with BJT input differential - I gaves lower THD, sonically I dont know now what happens.. I can only anticipate). |
|
|
| gnomus |
With out regulation is the Aleph 30 hum free? Or, is there normally some low level hum that can be heard at the speaker? Because, I remember reading some place that they work well with horn type speakers.
Still find the amp sweat, rich and full sounding. Very musical. And that works with music. Detail is lost. But, maybe over time the amp will break in and sound more open. I am not one that is in the camp that believes that electronics changes a lot after it does the "break in thing" . After an hour or so I feel that is the sound of the electronics and do not expect any big changes. |
|
|
| Nelson Pass |
Without supply regulation, but with selection of transistors,
we have been able to get the typical Aleph 30 down to
about 200 uV noise (20-20K unweighted). |
|
|
| gnomus |
Thanks!
200uV noise is hard to comprehend. I need an analogy. Like it would be as loud as the sound of an ant waliking on leaves if you could hear it. Maybe very faint hiss heard with your ear at the tweater is on target? Or a range. A high figuar that would be considered noisy and a low figuar that would be nearly impossible to hear. |
|
|
| Nelson Pass |
| I believe that would be about -83 dB referenced to 1 watt. |
|
|
| blank527 |
Hi,
This is a little off topic but can anyone tell me what the heat dissipation of an aleph 60 is?
On page 4 in the manual it says 240Watts.
Later in the manual it says 200Watts.
On the specs page it says 220Watts of power consumption
What to believe??:(
Thank you,
Remco |
|
|
| vpharris |
I think you're "splitting hairs" as we say.
How about 220 watts +/- 10% ? Is that satisfactory? |
|
|
| Nelson Pass |
| quote: | Originally posted by blank527
This is a little off topic but can anyone tell me what the heat dissipation of an aleph 60 is?
On page 4 in the manual it says 240Watts.
Later in the manual it says 200Watts.
On the specs page it says 220Watts of power consumption
|
I wrote all that. MuuuuHahahahaha....
:darkside: |
|
|
| Dennis Hui |
| quote: | Originally posted by Nelson Pass
I wrote all that. MuuuuHahahahaha....
:darkside: |
Maybe someone from the forum lucky enough to be going to
the CES can make a recording of Nelson laughing
his evil laugh. :)
Dennis |
|
|
| Algar_emi |
I Have distorsion quite easy to see on my Aleph30. Waveforms are ok at 1Khz, but a bump in the waveform can be see at soon as 10Khz, it is worse at 1Khz. Here the waveform, in 8ohms, at 10khz.
Any idea what can cause this distorsion and how can I remove it. One possible source, the AC gain feedback capacitor 220uf. I'm using a Panasonic FC type. Will it be better a Black Gate? What BG type?
Any suggestions?
Thanks. |
|
|
| Algar_emi |
| Here waveforms at 1Khz |
|
|
| Algar_emi |
| Here my Aleph30 on test |
|
|
| JDeV |
Hi Algar_emi
I also noticed similiar distortion when I tested my A30, but discovered that it was my Scope's probe leads that caused it. I don't think that the type of capacitor you use will change this problem, although higher quality caps will sound better in certain cases. BTW, can you hear this distortion, how big is this signal, is it on both channels? Did you check for possible bad solder joints? I used standard type Nichikon and Rubycon caps on my A30 and standard type Elna caps for the PSU and they all sound very good singing together.
Hope you get the cause of your'e problem and can fix it. |
|
|
| Henrik |
If You increace C8 from 10pF to say 20 pF You proberly would get rid of it the instabilit at the 10KHz squarewave, but on the other hand You also will lower the hig rolloff.
Try it out, but I think it is oky as it is. |
|
|
| Henrik |
| BTW really nice work!!!! |
|
|
| Koy |
| I tell you the same as Henrik did. 20pF is not so much Dont afraid. |
|
|
| Algar_emi |
The sound is very detailled, with a lot of air, but lean bass.
But all this only at low level. If I tried to crank to volume, the sound change and become congested.
I have a pair of acclamed Axiom 3Ti.
My preamp is a Bride of Zen.
The distorsion I mentionned is a higher listening volume, that why I think it explained th change in sound.
As described by others users, the aleph used the active Current source to increase efficiency. So, I think that the change in sound occurs when the active current source pull the amp in class A-B mode rather that pure Class-A. The Capacitor in the current feedback path is the on passing the AC and contributing, maybe, to the distorsion. By the way, both my channels are behaving the same.
Is a better FET matching in the Current Source could eliminate this distorsion?
I will tried to open the AC feedback an listen to it...
I'm waiting for any suggestions. |
|
|
| wuffwaff |
Algar,
I never heard such behaviour even with a 75% ac-current gain setting. Can it be that your amp is clipping?
This could be the case if your ac-current gain is chosen too low (under 50%) and your speakers demand the current but don´t get it.
william |
|
|
| Nelson Pass |
When you deviate from the 50% figure for current gain
on the current source, you get more distortion at the
highest output, either because you are current-starved,
or because the current source is shutting off on negative
peaks. The cure in both cases is more bias.
Come to think of it, more bias is almost always the cure
for any limitations of Alephs. ;) |
|
|
| Coulomb |
| quote: | Originally posted by Nelson Pass
When you deviate from the 50% figure for current gain
on the current source, you get more distortion at the
highest output, either because you are current-starved,
or because the current source is shutting off on negative
peaks. The cure in both cases is more bias.
Come to think of it, more bias is almost always the cure
for any limitations of Alephs. ;) |
Well we all have our Biases, some more than others. :) |
|
|
| Algar_emi |
Thanks Mr. Pass for participating to the discussion.
My current biais is 0.5A, as per Aleph30 service manual.
I adjusted the AC gain to 50% one week ago using the procedure described in one of the thread. I changed R21 from 820 ohms to 1K and got a perfect 50% gain.
My supply is using a 18+18V, 500VA transformer and the same bank of cap as the original Aleph30.
I tried using 0.8A biais but my output DC offset increased from the actual 120mv to more than 300mv! So I put it back to 0.5A.
This DC offset is causing me also a lot of problem. My input FETS are well matched. The Current and Active FET are almost identical also, except for one of the Current Source One.
Can I try to balance my Current FET by tuning one of the 221 ohms gate resistor on the FET that is different?
Mr. Pass you mentionned that R8 can be used to adjust the offset. What are the typical values to use? Do I need to used both R8 at the same time, or they are indicated at two different places?
Thanks for any comments. |
|
|
| wuffwaff |
Algar,
the 0,5A bias you mention, is this per fet or in total? The Aleph 30 has ca. 1.5-2A total bias and a 25 volt supply
Normally only the input diff pair is responsible for dc offset. If this changes so dramatically when changing bias I think something else is wrong.
william |
|
|
| Algar_emi |
The Biais is 0.5A each, total 1.5A per rail, 3A total.
What else can be wrong? Do you have any test that I can do to start finding the problem with the DC offset?
Both my channels have the same behavior. So I exclude bad soldering and bad parts. Possible causes then are:
bad schematic to start with, but the amp is working so...
Bad traces routing. The more I think, my problem is maybe cause by the way I connected the FET using point to point wire. They are connected like this (See picture):
+V +V +V
| | |red wire (all AWG18 wires)
Q6 Q7 Q8 FET Current Source (ref Aleph30 Schematic)
|___|____|0.47 ohms and org wire
|
| wht wire
|______________________
6 resistors 0.47 ohms |
| yel wire | wht wire
Speaker out + |
|-------|---------| org wire
Q9 Q10 Q11 Active FETs
| | | org wire and 0.47 ohms
--------------------
| yel wire
-V
I suspect That the single wht wire connecting both FET banks and the single wire connectiing the Active FETs to the negative rail are choking the current and cause current limitation in the circuit.
Any suggestion for the power traces and the speaker connection point |
|
|
| Algar_emi |
| Here a better schematic |
|
|
| Nelson Pass |
I don't think the negative supply leads are the problem
assuming the front end is also sharing exactly the same
negative supply, which is necessary.
As to offset, I assume you are using a cap in the feedback
loop. As your rail voltages are a bit low, I suggest that you
reduce the 4.75K resistor feeding the NPN positive bias
circuit to 3.3 K or so to compensate. This may in fact
cure your high power distortion.
Also you can adjust the Source resistance on the current
source which feeds the input diff pair to adjust the offset. |
|
|
| Algar_emi |
Thanks Mr. Pass. Your help is very appreciated.
I will tried these tonight.
You don't think my wiring can affect the sound?
I will rewire one of the channel so each current source feed one Active Fet using its own wire.
Just to be sure the rest of my circuit is Ok. Here the complete measurements:
Aleph30 Test Results
Supply Rails: +/-23.5V
Left Channel Right Channel
Diff Pair FETs Diff Pair FETs
3.751 3.745V 3.644 3.636V
offset 71mv offset 132mv
Current FETs Currents FETs
FET Vres Vgs FET Vres Vgs
Q6 .254 3.832 Q6 .245 3.735
Q7 .254 3.834 Q7 .248 3.727
Q8 .231 3.856 Q8 .241 3.729
Active FETs Active FETs
FET Vres Vgs FET Vres Vgs
Q9 .238 3.860 Q9 .220 3.947
Q10 .251 3.847 Q10 .246 3.920
Q11 .259 3.840 Q10 .242 3.923
Vres is the voltage accros the 0.47ohms on each FET.
Are the FET correctly MATCHED?
Thanks in advance. |
|
|
| Algar_emi |
| Here the Front End That I built. I hope there is no error in it. |
|
|
| Algar_emi |
After changing R4.75K to 3.3K as suggested from Mr. Pass (Thanks again Mr. Pass, you're the man)
my distorsion problem is present only near clipping, which is normal. According to the Aleph litterature, the active current source increase the Class-A efficiency. The clipping point is higher but distorsion increase near clipping. Exactly what I'm getting.
Max Power before clipping = 16.92Vrms in 8 ohms, 1Khz=35.8W
This is very good.
My dc offset problem seems also to be resolved. I'm will get the final results tonight after the amp is stabilized.
Everything is looking good.
In the mean time I will change the wiring around the FET to see if it improve the sound. |
|
|
| Algar_emi |
With the new 3.3K resistor for the current source of the input FET, I have near perfect offset on the right channel at <10mv.
The left channel is around 60mv. I will try to reduce it.
I re-arranged the wiring for the right channel. All FET have now their own wire and they all meet in a star fashion at the 6X0.47 ohms at the speaker output.
The sound improve with this mod. There is now more bass and it has more punch to it. I will do the same to the left channel. It probably also improve the DC offset at the output. The offset was higher before this mod. |
|
|
| Stanislav |
Hey, guys
Another question on this hot Aleph 30 topic -
where should one connect the R0 (the 3ohm@3Watt resistor) - should that be on the main board to ground ( and that means two R0, one for each channel), should it be on the PSU board (a common one, or two serarate ones for each channel), or something else.
Tell me how did you implement it, your input will be highly appreciated.
Stan - wants to be safe when possible |
|
|
| Algar_emi |
| You need two R0, one per channel. You need to connect close to the channel audio input gnd. Bye... |
|
|
| JDeV |
| quote: | Originally posted by Algar_emi
You need two R0, one per channel. You need to connect close to the channel audio input gnd. Bye... |
What is there purpose? What is downside of not using them? |
|
|
| Algar_emi |
It is to protect again ground loop. Use it. It doesn't affect the sound. Bye...
SB |
|
|
|
