John Curl's Blowtorch preamplifier part III

Status
Not open for further replies.
resistance. One of my 3V7 Samsung Lithium Ion cells delivers easily 30 A,
so don't forget your fuses or you might be sorry.
If you compute the source resistance from that: that is much less than
needed to explain the noise.

What happens when you suddenly short an unloaded battery can involve some non-intuitive things at least for wet lead acid batteries. It's been a while but I remember finding at least one analysis based on the details of the battery chemistry and what goes on at the electrolyte/plate interface. I do know shorting 8 - T105 batteries can shorten a screwdriver by 1/4".
 
Looks like he ignores being called out on those....:p

No, I just ignore statements that don't deserve a reply.

Frankly, I am surprised that you feel the need to respond like this, in support of someone who is obviously a troll and took my statements out of context just because he's tired of being wrong. I said absolutely nothing that is incorrect if you go back and read the context of the posts.

First, EDLC / supercaps are TYPICALLY higher ESR, and higher ESR than ALL MLCC decoupling caps as far as I can tell. Do you disagree with this?

Second , I never said ALL voltage regulators are quieter than batteries. There is ONE IC regulator family (LT304x) that may be. I also gave a link, TWICE, to its manufacturers measurements that showed it to be lower noise than a 3.7V lithium cell. Do you disagree with anything I just said?

So, I would appreciate if you were less careless in your post.
 
What happens when you suddenly short an unloaded battery can involve some non-intuitive things at least for wet lead acid batteries. It's been a while but I remember finding at least one analysis based on the details of the battery chemistry and what goes on at the electrolyte/plate interface. I do know shorting 8 - T105 batteries can shorten a screwdriver by 1/4".

We had someone short only a 2 Ah LiPo without a protection circuit with clip leads. The clip leads melted off their jacket and started glowing pretty much instantly. Fortunately someone knocked it off the bench quickly and it opened itself before the battery went up in smoke.
 
Anyone ever look at the limit of 1/f noise? Try setting the center screen trigger on a digital oscilloscope so you can look at a peak AC voltage of around .1 volts on your power supply. How long does it take till it triggers? Then try looking for 1 volt. Depending on luck you might capture one of those from waiting overnight.

A valid design consideration is how often the 1/f noise might actually cause a spike that could be perceived in the system. The follow up is how does the rest of the system behave to such noise. Almost never an issue with even modest PSSR and a few small local filter capacitors.

But then there are some audio designs that just might not handle 1/f vvlf noise well.
 
Last edited:
No, I just ignore statements that don't deserve a reply.

So @Jakob2's specific request to you was something that falls into that category? That's what I was referring to here, I admit somewhat obliquely. If you missed it and need a link, I'll find it.

The irony here is you're excoriating @mmerrill99 for displaying his ignorance here but in the context of @Jakob2's request you prefer to keep your own under wraps. Are you saying that ignorance shouldn't be on display?

Frankly, I am surprised that you feel the need to respond like this, in support of someone who is obviously a troll and took my statements out of context just because he's tired of being wrong. I said absolutely nothing that is incorrect if you go back and read the context of the posts.
There's a need? Just a spur-of-the-moment decision to highlight your apparent (to me) inconsistencies. You're reading what you wish to read into my post, I'm by no means supporting @mmerril's technical arguments against you. Its by no means obvious to me he's a troll, but its clear he does enjoy trolling you.
First, EDLC / supercaps are TYPICALLY higher ESR, and higher ESR than ALL MLCC decoupling caps as far as I can tell. Do you disagree with this?
No, I don't agree with them having higher ESR than ALL MLCCs, that's a claim way too far in my estimation. But in principle I do agree that EDLCs have high ESR, relative to their capacitance when compared to MLCCs and electrolytics.


Second , I never said ALL voltage regulators are quieter than batteries. There is ONE IC regulator family (LT304x) that may be. I also gave a link, TWICE, to its manufacturers measurements that showed it to be lower noise than a 3.7V lithium cell. Do you disagree with anything I just said?
I haven't done the background research into noise of regs vs batteries to comment either way.

So, I would appreciate if you were less careless in your post.
If you could highlight the carelessness which is apparent to you then I'll endeavour to address it.
 
I have no time for Jakob and his games. Really.

Yes, obviously comparing a tiny valued MLCC and a huge electrolytic cap isn't fair. Hence I said typically. To be even more clear, the original post where I made that statement was after discussion of a 50 ohm ESR part. Again, having your context right helps.

Since you didn't do any research on the regulator or battery noise, then how would you know what is correct?

I said you were careless because you didn't even read my posts to make sure that I said the things mmerrill99 claims. If you did, you would know that I did not say that voltage regulators in general are lower noise than batteries.

Might have to finally break out the ignore list after 15 years.
 
I have no time for Jakob and his games. Really.

Clearly that's not so or you'd not have time to notice his posts - he'd be in your ignore list. You also found time to post a claim that he'd been posting 'nonsense'.

Since you didn't do any research on the regulator or battery noise, then how would you know what is correct?

Right, I don't know nor did I claim to know.

I said you were careless because you didn't even read my posts to make sure that I said the things mmerrill99 claims. If you did, you would know that I did not say that voltage regulators in general are lower noise than batteries.

As I've already pointed out, I'm not supporting @mmerrill99's technical claims against you. You're normally factually quite correct in respect of technical issues. Personnel issues, not so much.
 
I know what you are thinking, but that is not what happens with a supercap.
The early series of supercaps intended for memory/presets retention were relatively high esr and this could constitute damping of DAC supplies.
Joe/mmerrill99, which makes/model of supercaps have you experimented with and what subjective changes have you noted ?.
Joe, what do you think is happening ?

Dan.
 
Last edited:
It's very hard to extrapolate, since the result depends on the amplifier load (current through the coil), output power, overall 2nd harmonic distortions (if that's rather high, you will barely see an improvement), coil diameter and length, amplifier wiring, etc... But sure, attached is the distortion spectra with the output inductor off board. After removing the inductor, the 2nd harmonic went down by 19dB (almost 10x less 2nd harmonic, at this very low level).

3 pairs of lateral mosfets, biased at 150mA each, 100W/8ohm, coil is 15mm dia and 25mm long (2.5uH) amortized by 3.3ohm, frequency is 20KHz, THD @20KHz with coil on the board 0.001%, coil on board, and 0.0001% with the coil off board (as shown). It is expected the on board coil effect to scale more or less linearly with the current, so with the square of the output power.

This was an extremely careful construction (the PGP) with regards to stray magnetic fields and induction (bringing a steel screwdriver around the on board coil was immediately visible in the distortion spectra, as a 10-15dB extra increase in the 2nd harmonic distortion). I can't even imagine what would be the effect for a conventional amplifier wiring, since I never placed an output coil on the board since, not even for much more relaxed YAP amp.

P.S. Found another pic, the 20KHz distortion residuals, with the coil off board. As you can see, almost entirely 3nd harmonic. 2rd harmonic due to magnetic effects are the small ridges in between, placed at the load current peaks.

Thank you for your effort.
Do you have information how much the value of the coil inductance change that influence? I use 0.5uH up to 1uH as I try to design an amp to be stable on capacitive load with that values.
 
Unfortunately not. All I can tell is that the mutual coupling factor is inversely proportional with the number of turns.

Speaking of this, find attached a photo of the innards of the Halcro DM88 amplifier (the lowest distortions commercial amplifier ever build, as far as I know). Not only do they have the coil off board, but they placed it in a special closed (and I suppose shielded, check out the walls thickness) compartment of the chassis. Guess why they bothered to do this.
 

Attachments

  • B2200AF0-D6B9-4113-813C-C739FC10479E.jpeg
    B2200AF0-D6B9-4113-813C-C739FC10479E.jpeg
    89.1 KB · Views: 200
Last edited:
THD @20KHz with coil on the board 0.001%, coil on board, and 0.0001% with the coil off board (as shown). It is expected the on board coil effect to scale more or less linearly with the current, so with the square of the output power.

Frankly, do you find efforts like this any useful, other than engineering feeling of killing some barriers? No one would hear non-linear distortion below 0.01%, provided it is not a cross-over like distortion with spectrum to infinity, and even then it would be difficult to tell it by listening.
 
Thanks vac!

Is there phase adjustment in dsp? I read it held zero phase through the outputs no matter settings but didn’t know if it could be manually manipulated in dsp?

Also is the volume part of the analog out or is that fixed....in other words could it be used as a preamp?


I am not aware of a phase adjustment (delay?) but why would one need that? The volume control is done in the digital domain, so yes, it can be used as a preamp. I intend to use mine also for vinyl playback once I get all the other bits and pieces together. Now, its main use is USB and Toslink playback, when it is not being used for measurements.
 
Status
Not open for further replies.