Help with transistor substitutions

[trwh]

Hi

I am gathering the parts to build an amplifier design which contains a few obsolete bipolar transistors. They are the 2SD756, 2SA872, and the 2SB716. I have to hand the specifications of these devices, but I am having difficulty matching more modern devices to the older parts’ specs. (As a shortcut, if anyone has experience of these parts and can suggest alternatives, that would be much appreciated!).

Anyway, the problem I have is that for instance, the 2SD756 has a typical Hfe of 700 - yet in my supplier’s catalogue the devices are each specified with a range of Hfe values, typically say 200-500. I cannot find any that have a minimum guaranteed Hfe of 700, would any range of values be acceptable as long as 700 is the mean value? How critical is the matching of obsolete parts? This circuit has a MOSFET output stage, which makes me worry that the amp will oscillate if the bipolar driver transistors aren’t properly matched to the design.

Thanks in advance for any help,
Tim.

 

[JensRasmussen]

Transistors

This schould work

2SD756 = 2SC2240

2SA872 = 2SA970

2SB716 = 2SA970

If you can't find them, there are more substitutes avalible

Good luck

\Jens

 

[trwh]

Jens,

Thanks for the substitutes!

I can get the parts you mention. I notice that the ft is significantly lower in the 2SA970 (120MHz), than in the original 2SB716 (350MHz). Is it likely that this parameter will affect the operation of the circuit?

Thanks again,
Tim.

 

[JensRasmussen]

FT

Hi Tim

Well It's hard to say, I would think that the transistors are in the inputstage (Do you have a schmatic?)

I'd try it out, powering it up with a scope on the output making sure it's not oscilating.

Other transistors:

2SB716 = 2SA970 = 2SA1123, 2SA1019,2SA1281

\Jens

 

[trwh]

Hi Jens

Unfortunately, I cannot find a source for the other alternatives you mention. However, I will try your original substitutions and check for oscillation - thanks for the info!

Will any transistor with a prefix 2SA and same polarity / similar figures be suitable? I may have a closer look at more devices.

Here is the schematic for the amp, TR6 is 2SK135 and TR7 is 2SJ50. The amp can produce 100W/8ohms, 150W/4ohms, and is said to be very reliable, although not really a HI-FI design.

Thanks again for your time,
Tim.

 

[JensRasmussen]

Transistors

__________________________________________________
Will any transistor with a prefix 2SA and same polarity / similar figures be suitable? I may have a closer look at more devices.
__________________________________________________

They don't have to be 2SA, 2SC or what ever, these names are mostly japan og at least asian devices, european small signal devices are often called BC.

\Jens

 

[mcp]

Hi trwh

Judging from your schematics, there is one output pair of lateral mosfets (2SK135/J50). I have not worked with these for quite a while as they are now obsolete. The newer replacements are 2SK1058 / 2SJ162 (plastic pack type). I believe they are very similar. Nonetheless, concerning power output, the best I could get out of 1 pair of 2SK1058/J162 is:

8 ohms/4 ohms/POWER FACTOR/Vs+-/Transformer
45W/80W/1.78/+-33V/25-0-25 VAC
60W/80W/1.33/+-40V/30-0-30 VAC
80W/90W/1.12/+-46V/35-0-35 VAC
(RMS POWER into resistive load @ 1KHz Sine)

Regarding suitable replacements for the first and second stage, I believe your concern is whether the amp will oscillate if other part numbers are used.

Even without any replacements, just from the schematics alone, I would say that there is a high probably of some oscillations. This is due to R13/14 at 100R. To remove the parasitic oscillations at the gate, we are looking at something like 500R. Of course, this will be at the expense of some bandwidth.

Technically, there is nothing wrong with using 2SA970 and SC2240. However, I would feel more comfortable with a slightly beefier trans for the 2nd stage, more towards the 1W type. Also note that the 970 and 2240 are Low Noise trans. Good for specs, get very impressive measurements, but personnally, it dulls the music. I would rather restrict these trans for the first stage only. I use Low Noise transistors very sparingly, but I would not hesitate using them for a discrete mic pre-amp when I am looking for a gain of 100dB with as low a floor noise as possible.

You may like to try these Motorola part no for the second stage: MPSW?? or the MJE340/350.

Regards
Mike

 

[trwh]

Hi Mike

Thanks for your sharing your experience! I have taken on board all of your advice – I shall investigate the alternative output devices, and larger second stage transistors.

The article that accompanied the schematic above did mention curing oscillation problems by increasing the value of R13 and R14. Any value between 330 ohms and 1k ohm was possible. They also suggested placing small value (47-330pF) polystyrene capacitors between the gate and source terminals when using the higher value gate resistors, to help maintain bias current during signal peaks.

Before I commit myself to buying parts for this amp, does anyone else have any suggestions for a very robust and straightforward amp that can deliver 100W / 8ohms? This is one of the only simple, yet still relatively bombproof designs I have seen.

Thanks for your help,
Tim.

 

[daatkins]

Hello Tim,
I suggest you try Grandata of Wembly for the missing transistors:
http://www.grandata.co.uk
(you need to click on catalogue -> semis -> jap transistors). I've used them before and they offer good service.

As for the output MOSFETs, Maplin have been substituting Exicon 10P16 and 10N16 for the last few years.

Regards,
David