Aleph asymmetrical clipping? (help)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I am veru much aware that the topic of assymetrical clipping of aleph amps is something overtalked, but I am really confused now after my second attempt of building an aleph. What I mean is that my new aleph30 biased at 1,5A +/-27,5V(680W torroid 200000uf total) clips assymmetrical on the lower side of the waveform at 50% current source setting. To correct this I have to raise this setting to 66%. :confused:
Power output in 8ohms is about 30W clean and 40W clipped.
However during my recent hearing tests on my TannoyHPD385 as well as on a back loaded horn with fostex fe208 sigma, confirmed my initial impression. The sound is somehow empty and annoying, bass is o.k but mids are very forward and harsh and upper treble is almost dark. The comparison was done using a DIY PP6550 tube amp, preamplifier was a very good opa627 and source a sony xa50es at 3A setting. I really do not want to believe that a simple tubed PP outperformed by a huge margin my beloved aleph. :bawling:
I hope there is something wrong with my setup around the current source. Maybe the amp current setting should be left at 50% and assymetrical clipping? I really do not know but I will continue testing and let you know about the oucome. Meanwhile any help would be very much appreciated ! :mad:

(all mosfets are matched to 10mv and offset is 5-10mv on both channels)
 
panos29 said:
What I mean is that my new aleph30 biased at 1,5A +/-27,5V(680W torroid 200000uf total) clips assymmetrical on the lower side of the waveform at 50% current source setting. To correct this I have to raise this setting to 66%

So, to be clear:

The positive side of the waveform is OK, but the negative side is clipped?

Looking at the Aleph-30 schematic on passlabs.com, I can imagine the following might be a problem:

- insufficient -ve supply voltage (e.g. rectifier diodes badly connected)
- wrong value source resistors R37-39
- insufficient current through Q3 etc, giving insufficient drive voltage across R14, or R14 wrong value
- current limit circuit (R15/R16/Q4) operating prematurely - wrong resistor values?
- input protection diodes Z1, Z2 connected wrong way round!


I assume you've got a scope to look at this. When it's clipping, can you measure the voltage between gate and source of Q9-Q11, and between drain and source of Q9-Q11? This might give us some clues.

Cheers
IH
 
Panos,

normally when raising the ac-current-gain you reduce the clipping only on the positive half of the waveform because you raise the maximum positive current.
The maximum current on the negative half is only limited by (R15/R16/Q4) wich is normally set in a way that the amp will always clip first on the positive half. In fact you can leave these out if you´re shure not to shorten the outputs of the amp.......

william
 
measurements

o.k I finally did some measurements as requested,
so, we have Gate Drain voltage 23,6
Drain Source 26,9
Gate Source 3,3
Voltage on R14=3,5
Voltage on R11=5,44
V+=V- =27,6V
protection diodes are correctly positioned and only on the positive side as negative is grounded (I use single ended pre). Protection circuit around Q4 is in off state Vbe=0,123V on clipping.
Clipping voltage is 17VRMS on 7,5 ohms, all measurements taken on the beginning of clipping. The clipping is firstly evident on the POSITIVE side (muffwaff is correct)and not on the negative as I stated. Moreover all measurements were done with 52% current setting. All values are exactly as in the schematic except the R19 which had to be 22K in order to achive 0,25 on each 0,47ohm source resistors for 1,5A bias. After much experimentation I noticed that R8? in parallel with c11 helped to make the clipping more symmetric with tha value of 4,7K.
 
Hi Panos,

The Aleph must have assymetrical clipping by design.:bigeyes:

It will always clip first on the positive side because that is where it will run out of current first. The negative clipping you see (if any ) is only caused by the negative current limiter around Q4.

So if you want symmetrical clipping (why?) you´ll have to change one of the resistors around Q4.

17 volts at 7.5 ohms means around 38.5 watts and 2.26A rms and 3.2A peak. With 1.5A bias this means an ac current gain setting of around 53% So this also fits.

It seems that your Aleph is working allright then.........

Please explain why you would like it to clip symmetrically?

William
 
Dear William,

Thank you very much for the help, I just realized that assymetrical clipping is an inherent feature of single ended amplifiers! So, I guess my amp works o.k and I am happy for that as much effort was put in the mechanical construction of such a project. However the reason that I thought there is a problem with assymetrical clipping is the sound impresion I got out of this amp. As I stated in my first post, the sound compared to an PP6550 tube amp is not bad but definitely inferior in most ways. Namely there was no body to the music and upper treble seemed very constrained and limited, moreover the mid region (women voices) were annoying and very forward like there was an amount of distortion introduced and in general the sound was tiring and not relaxed as with my tube amp. My speakers are very easy load, very eficient and fairly costant load through their bandwidth. Of course the test was performed with current gain setting at 66% and I hope that this was the reason for the bad outcome . I will test it again within the week and let you know about the outcome.
 
Ðåñßåñãá óõìðôþìáôá- ìçðùò êñýùóå ôï ÁëåöÜêé?

Seriously though there shouldn't be that much of a difference between the two amps.Did you ever compare your Aleph to another amp originally or is this the first time? If all working voltages check out and input output connections are correct you have a dilemma there.How big are the differences you describe?
 
Ðïõ íá êñõþóåé ìå ôüóç æÝóôç ðïõ êáíåé áêüìá åäþ? :hot:

So, since you ask, my aleph was seriously ill judging from the sound. Even a "stupid" gainclone I use for my computer speakers seems better especially at the upper treble region! My pp6550 utperforms by a huge margin my aleph in every aspect. The sound is "cold" and clinical and not at all full bodied and I was not the only one in the listening session who felt that way. I have to admit though that the amp had very low "mileage" about 3 hours from "new", so there may be some improvement. I really hope it will get better, since I set the current source at 50%. if not...:smash: I am really concerned about the final outcome as I allready got all the parts including heatsinks for the aleph-X, and if the sonic signature is similar I will abandon it and go for a PP813 tuba amp project that I always wanted...Time will tell!
 
I think there must be something wrong.. my Aleph 30 is clearly better than any amp I've heard on the mid-hi region, it only lacks some power on the bass area, but with only 30W ....

With the values on the schematic included in the service manual my amp idles at about 2.1 A, if your heatsinks allow you I'd raise the bias to at least 2 A.

Now I'm going for a Aleph-X also, when it's finished I'll be able to compare them.

Cheers

Andrea
 
Mçí åãêáôáëåßøåéò ôéò ðñïóðÜèåéÝò óïõ áêüìá.

É used to have a pretty good Jadis Orchestra tube (PP w KT90) not to mention a couple of Naim amps before my Aleph 5's.
So I did have something good to reference them to and as I said before the Aleph should be at least in the same league if not better in some respects.It is a pretty smooth sounding amp and should not have mid-range problems where a lot of it's strengths are.Soundstaging and depth should also be very good.The only aspect of the Aleph which could come under relative criticism compared to certain other hi-end amps may be a slight lack of bass speed but this is depending on implementation(bias/caps/bypassing etc), partnering equipment and so forth and only on points not in a very obvious way.In fact I find that the Aleph is a pretty easy amp to live with leaning towards the warm side of neutral rather than hard or analytical so if yours is not sounding this way perhaps something is wrong.
The source of your problems may not be clipping per se but perhaps a semi faulty semiconductor somewhere.Can you compare the sound of each channel separately on the same speaker and with the same channel source and check if they sound the same.If they do sound the same then a faulty component is probably not the problem but it could be something in your implementation.In my view burning in should help a new amp but not that much.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.