neve preamp and high gain transistors

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

rafafredd

Well-known member
Joined
Jun 3, 2004
Messages
2,409
Location
Rio, Brazil
I´ve got some bc184L transistors and they measure 250-350 hfe.

I´ve read that the neve 283 preamp and output stage calls for 600Hfe or more, hence the need of bc184 "c" version.

But do I really need that high of a Hfe for acceptable noise levels? Does someone in TheLab tried lower gain transistors, like 350Hfe on this circuit?

As always, Thanks!
 
you are right. I´ll have to cross some legs there... I hate it.

But anyway, do you think it´s usable with 350HFe?

I may add that it´s for mic inputs...
 
neve_b1.gif


> I´ve read that the neve 283 preamp and output stage calls for 600Hfe or more, hence the need of bc184 "c" version. But do I really need that high of a Hfe for acceptable noise levels?

Hfe matters to noise, but really the square-root of Hfe. 600 or 350 will not make an audible difference unless the source impedance is very high (high step-up transformer).

It can also affect DC bias levels. Personally, I think any circuit that won't bias-up to about the right place with beta=50 or maybe 100 is just reckless. It appears the Neve designer felt the same way: you'd have to get down below beta=100 to see any bias-shift larger than resistor tolerance, and I think it would bias just fine (slightly off nominal values) with beta=50.

Just build it.
 
Great!

I´ll be using a 1:2 input transformer, like the original 10468, so I think I´ll not run into noise troubles!

Thanks PRR!
 
I'm not sure about Brazilian availablility, but in my hunt for building stuff out of commonly available parts as one of the prerequisites, I settled on the very common MPS-A18 as my high gain NPN of choice.

The HFE has been quite consistant, as has the low noise performance over the thousands I've used.
 
Well, yes, I have some of those here.

Do you think I can use it there without problems?

All of them really tests very high HFe, typically greater than 600!
 
I have been using the MPSA18s everywhere they call for NPN input transistors (2N5088, BC550, etc.) and they work really well. Higher hfe on the input will not hurt. Start soldering something to something...
 
I built a two stage Neve mic pre thing (BA283AV) and while at the highest gain setting (~70dB) I noticed that it was quite noisier than say the Aurora GTQ2 (at a comparable gain) I had in the next rack. I used BC184C transistors from Newark, but I didn't check any for hfe or anything. Is there a replacement transistor that might be quieter? How would I check hfe with what I got (Fluke meter, couple power supplies, o-scope)?

Thanks,
Ian
 
There is no recommended PNP counterpart to the MPSA18. High hfe PNP transistors that may pair up are 2N5087 or PN4250(A).
 
[quote author="Dan Kennedy"] I settled on the very common MPS-A18 as my high gain NPN of choice.[/quote]

Just thought I'd bump this bit for those that weren't paying attention.
I have a Great River and obviously have had the lid off.

side note,
has anyone every tried a TIP unit in the power output stage as a replacement for the TO3 package ?
 
Yep, TIP41C, works great.

Mounts to a nice little 1 1/2W heatsink from Aavid, Thermalloy and most everybody else that makes extruded heatsinks.

A little clearer sounding than the normal 2n3055 implementation, at least to me.

But ground plane circuit board layout, local supply regulation, different transformers all add up to make the differences too.

Again, the TIP is an available, commonly used industrial part. Means it won't go away too soon.
 
[quote author="Dan Kennedy"]Yep, TIP41C, works great.
...A little clearer sounding than the normal 2n3055 implementation, at least to me.[/quote]

thanks Dan,

I was actually thinking of the TIP3055.
My local has these on the shelf. I'm still working on the Project 1 PCB. A simplified 1272 board directed at DIY. Not trying to be a 1272 clone but a rework using run of the mill transistors and a simple build. Having many options on transistors is a good thing if the bias and balance can be kept right.
Then add the JLM switch kit to extend the gain.
It will happen one day.

All this without getting in the way of some of the commercial products out there.

oh Dan ... love that interview
:thumb:
 
I posted in the BlackMarket, but I guess it's better placed here in TheLab:


Just wondering a bit about the transistor-types...

For the BA283-cloned PCB-cards, to keep this in style I figure
it'd be 'the right thing' to use those BC184C NPNs.

But do I understand it correctly from the JLM-info that Neve
themselves also used the types as mentioned here: http://www.jlmaudio.com/neve_BA283.htm ?
(not sure about the interpretation of 'original component variations')

Of course the influence of all this will be really a lot & lot less compared
to the TX-choices etc etc, but just for grins I want to keep close to the
original stuff for these cards.

For Fabio's relayouted cards it'll be another story of course - what are people using there ? MPSA18 ? BC550C ?

Bye,

Peter
 
I'm using BC549C all around for the Neve stuff... first transistor in ea stage I choose for 600 hFE... :thumb: Sounds good to me.
 
Rob et al
May have been asked many times...
We are looking for a hfe over 600 so... anybody got the best circuit for testing this with the minimum of components
 
You really want to test the devices under similar bias conditions as the circuit. At least find out what the nominal collector current is and what the collector emitter voltage is, about. You could then determine what the base current is by measuring the voltage drop across a base resistor after putting a resistor in series with the emitter to "ground." Tie the collector to positive, and run the base from a voltage divider of moderate impedance to set the emitter current to the nominal operating point of your end-use circuit.

Example: suppose operation is at about 5V c-e and 1mA. +10V power supply, 4.3k emitter R to ground. Voltage divider set up for half of the 10V before the transistor is connected. For a beta of 600 the base current will be 1/601 of the emitter current, hence about 1.66uA. If the impedance of the voltage divider is 50k (each R 100k), then connecting the transistor will change the voltage from 5.00 (or whatever it is exactly) to about 83mV less (assuming this is set up for NPNs). Any smaller change means the beta is higher than 600.
 

Latest posts

Back
Top