Reducing gain in Hiraga pre-pre

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Member
Joined 2005
Paid Member
I was asked if it's possible to reduce the gain of a Hiraga pre-pre. The reason for doing this is that the gain of the following stages is too high (could reduce the gain there, but....)

I attached the Hiraga pre-pre schematic for those who are not familiar with it.

Any suggestions, ideas, comments?
 

Attachments

  • prepre_c.gif
    prepre_c.gif
    10 KB · Views: 299
Why change the current mirror? That is just a bias source, with perhaps some degree of thermal tracking. Changing the collector resistors will shift the DC collector voltage but that doesn't matter as the output has coupling caps (although no ground leak resistors).

The gain is just gm x Rc, so halving the collector resistors halves the gain.

Ummm... And gm=Ic/Vt, so decreasing the collector current also decreases the gain. Now the bias current is (Vc-Vbe)/Rbias, and the voltage in the collector is Vc-Ic*Rc=Vc-(Vc-Vbe)*(Rc/Rbias) and the gain is (Vc-Vbe)/Vt*(Rc/Rbias). Therefore, if you keep the resistor ratio constant, the gain and the collector bias voltage won't change.

FWIW, this schematic, implemented with real life discretes, sucks in any way I can think of. For example, the common mode gain has zero dB PSRR, and any inherent asymmetry in the differential mode will amplify the common mode crap. Thermal tracking of four discrete devices is virtually impossible. Current through the cartridge is zero only if the gain transistors are perfectly beta matched and all four transistors are perfectly thermal tracked (not gonna happen). Etc...
 
Last edited:
Waly said:
FWIW, this schematic, implemented with real life discretes, sucks in any way I can think of. For example, the common mode gain has zero dB PSRR, and any inherent asymmetry in the differential mode will amplify the common mode crap. Thermal tracking of four discrete devices is virtually impossible. Current through the cartridge is zero only if the gain transistors are perfectly beta matched and all four transistors are perfectly thermal tracked (not gonna happen). Etc...
Yes it is a spectacularly bad circuit. I was too polite to point this out.

I suppose this means one or both of:
1. the designer is a recognised 'guru'
2. commercial versions of it are hugely expensive to buy
 
Yes, perfectly awful. :eek:

Presumably the input impedance depends on the transistor's beta, so will be highly variable.

With a cartridge connected, there's probably some very significant asymmetry in the frequency response of each channel due to this.

Incidentally, has anyone done any research into the effects of bias current running through magnetic cartridges?
 
monty78pig said:
Presumably the input impedance depends on the transistor's beta, so will be highly variable.
No. Emitter input impedance is just 1/gm so is set by current alone. This will be quite well balanced as it is primarily set by the negative supply rail voltage and emitter resistor values.

I think the general consensus on DC through cartridges is "Don't do it!".

Incidentally, this preamp can only be used with low to medium output MCs. For a BJT the %THD is about equal to the peak mV input at the base-emitter, so only SET fans will wish to put more than about 1mV into it.
 
Administrator
Joined 2004
Paid Member
Yes it is a spectacularly bad circuit. I was too polite to point this out.

I suppose this means one or both of:
1. the designer is a recognised 'guru'
2. commercial versions of it are hugely expensive to buy

1. All the above are true,
2. I actually like the guy,
3. Despite the truth of all technical comments in this thread it actually works
well and sounds quite good. (I've heard one)

Personally I prefer Lundahl amorphous core SUTs like the LL1941 to pretty much any other step up device I have heard, so my comments are of course suspect. All of my current LOMC have source impedances of 6 ohms or less. (A bunch of SPUs and a Windfeld)

Was thinking of trying 6C45 in parallel for a head amp design, but never got to it. I thought noise might be a problem and that demotivates me.

Friend built me a multi-tan head amp which I use with certain cartridges like DL103D where their comparatively high source impedance makes SUTs more problematic.


Hiraga is very DIY friendly.
 
Hiraga is DIY friendly? :p

I've never seen any more fiddly circuits (especially in terms of quiescent adjustment).

He also provides exceptionally poor value for money in his designs, all things taken into account. Simplicity only in terms of electrical component count, and even then, it's very dubious.

Perhaps you should be on DIYAUDIOPROJECTS ;) , this is DiyAudio :D !
 
Administrator
Joined 2004
Paid Member
Spoken like a true expert Monty. I figured you might be a little more open minded, and I am sure you have heard and built many of his designs to know just how fiddly and difficult they are to get running. :D

Everyone is entitled to their opinion, if you like overly elaborate marvels of technical perfection that's just fine, but you might be surprised to discover that simple and flawed things often work as well at a given task if the designer understood what was actually important. Complexity as an end goal isn't good engineering. You might want to read up a bit on Hiraga, and then feel free to disagree with me, at least then you will be talking from a position of knowledge.

Last time I checked matching transistors wasn't that difficult, thermal tracking is far from ideal, but how much perfection do you need?

And no it wouldn't be my choice either.
 
Truly simple circuits are those with the right number of components to do the job. Less than that and things get complicated because you have to rely on tricks.

Exactly. I'd rather solve a problem with a transistor or two, than have to adjust a preset.

As far as I'm concerned Hiraga is the initial perpetrator of the subjectivist cancer that now riddles audio design. Douglas Self has a good article on his site that chronicles this decline. Hiraga features as the centre of that decline...

I strive for simple circuits. I've nearly finished Active Volume Controls an article on a new active volume topology that is simpler and better than the conventional circuit, as an example.
http://www.objectivesounds.co.uk/articles/active-volume-controls/
 
Administrator
Joined 2004
Paid Member
Wow, we really are on diametrically opposite poles in some ways.. LOL I found Hiraga entertaining and eye opening when I was introduced to his writings 30yrs ago. I suspect I get a lot more enjoyment out of audio today as a result. His writtings (in French anyway) had a technical bent missing from today's discussions. Hiraga introduced the West to Japanese audiophile sensibilities.

In the U.S. you would have to blame J Gordon Holt: https://en.wikipedia.org/wiki/J._Gordon_Holt - certainly more influential here than Hiraga who remains a fairly small blip in the English speaking world. Oh I enjoyed Holt's writing as well.

I would go a bit further, and state I have no problem with subjectivism unless it is a mask for ignorance of fundamental science which it so often is. Appliance operating well healed and often obsessive audiophiles drive me nuts.

These days I design medical electronics rationally for a living, but enjoy a different approach to my audio design. Once upon a time I was part of a design team that designed really high production volume consumer audio products, (Millions made ugh) later still I designed MI electronics. You definitely would not appreciate my current design approach where listening does play a big role in the design process. A few people here and there appreciate the way I do things, but not overly many. I have a well equipped lab and an obsession for measuring things, but I don't design the way you do or for that matter many around here. You might be right maybe I would be better off somewhere else, OTOH I've been here much longer than you and you're stuck with me for reasons that should be obvious.. :D
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.