Leach amplifier MPS8099/MPS8599 replacements

[john78]

Hi

I am building the Leach Amplifer. The Motorola MPS8099/MPS8599 are obsolete, can I use 2n5551/2n5401. Or would BC556 or KSC1845 be a better choice for replacing Motorola MPS8099/MPS8599? Has anyone built this amplifier with transistors other than Motorola MPS8099/MPS8599?

 

[Mark Johnson]

Mouser.com has both of them in stock . . . . . . manufactured by Central Semiconductor rather than Motorola

(link one)

(link two)

 

[john78]

NO kidding...I do not need help cutting and pasting into google search. Your inane reply answers NONE of my questions.

 

[Disco-Pete]

Your statement about obsolescence is rather suggestive. Patience, as I'm struggling with, is, well..

worth it. 🙂

 

[wg_ski]

I have always used MPSA06/56 interchangeably with 8099/8599. I also don‘t see any reason why the 5551/5401 couldn’t be used in the Leach amp either. The input stage is degenerated in that design, therefore the exact characteristics matter less than if it wasn’t.

Major makers are also dropping the A06/A56 in thru hole already, but SMD versions will continue to be made. No idea about current stock levels, though.

 

[Mark Johnson]

I suppose you could attempt to replicate my success from the year 2016: find a trustworthy seller of N.O.S. goods and buy some original Motorola / OnSemiconductor ("ONS") devices as New-Old-Stock.

Here's the order I placed. Pretty clearly they enforced a minimum line-item price of USD 100 or greater, except that I bought their entire supply of 2N5401 and it only came out to $42. I think they may have changed the line item min amount since then (?)

Date Ordered: Monday 14 November, 2016​

Shipping Method: FedEx Ground​

--------​

​

PRODUCT INFORMATION​

​

Part: 2N5401 (ONS) Qty: 1417 pcs. Price Ea: $0.03 ea. Line Total: $42.51​

Part: 2N5551BU (FAI) Qty: 3334 pcs. Price Ea: $0.03 ea. Line Total: $100.02​

Part: MPS8599RLRA (ONS) Qty: 1667 pcs. Price Ea: $0.06 ea. Line Total: $100.02​

Part: MPS8099RLRMG (ONS) Qty: 1429 pcs. Price Ea: $0.07 ea. Line Total: $100.03​

Part: BC550CBU (FAI) Qty: 3334 pcs. Price Ea: $0.03 ea. Line Total: $100.02​

Part: BC560CTA (FAI) Qty: 3334 pcs. Price Ea: $0.03 ea. Line Total: $100.02​

I've still got quite a few of those devices and stored them in the original shipping bags; the photos below show the date codes were 2005 and 2004.

I got lucky; maybe you will, too. (??)

_

 

[wg_ski]

If the entire stock times the unit price comes out less than $100 then they couldn’t enforce the line item minimum, could they? If that were the only item you ordered, they might have just raised the unit price to make the invoice $100. You will never know now.

I probably would have still ordered even hundreds or thousands of pieces, even if it came out to $120 for a batch, for the ones where there was sufficient stock.

Three of my favorite surplus outlets have folded since the pandemic. They were my usual sources for 2N/MPS jellybeans. My current parts for that class of device is the KSC1008/A708 - the EBC version (It comes in both EBC and ECB) because I could get them at Mouser. Still have a few dozen A06/56 left, and one or two 8099/8599, but new design has switched over.

 

[john78]

Wow any idiot convention. This is EXACTLY why most people leave DIY. If you have NO INFORMATION on the subject matter or technical information with actual experience, DO NOT POST IN THIS THREAD!

 

[GlennW]

I built a Leach using KSC1845 and KSA992. Works fine. Be careful, these are ECB.

Glenn

 

[john78]

Glenn, could you elaborate on:
1) Was this the 4.5 version of the leach amp
2) what rail voltages did you use
3) how did you fit the KSC1845/992 in
4) did you make the boards or use boards from Mr Leach
5) how long ago did you build the amp and how long has it been in service?

 

[Mark Johnson]

The OnSemiconductor datasheets I have, don't include all four of the h-parameters. In particular, the parameter h_oe =(delta_vce / delta_ice) is omitted, regrettably.

So I measured h_oe on a bunch of transistors, hoping to find which was the best and which was the worst. Here's the raw data for the MPS8599 and for the 2N5401, which often appear in the same positions within audio amplifier circuits. As you can see, the 5401 has got much better (lower) h_oe , therefore much higher Early voltage, which some designers greatly prefer.

{ Horizontal axis is |Vce| in volts, vertical axis is |Ice| in milliamperes. The curve tracer software insists upon displaying all plots in quadrant I. }

Best of the best measured h_oe values was the ZTX757, but be careful: it is device-optimized for a different region of the (Ccb, fT) space than these two.


_

 

[wg_ski]

5401 may be a better transistor, and C1845 better still. But how much resistance is actually sitting in the collector circuit on the diff pairs of the Leach amp? Not much, IIRC. And there are emitter resistors which further increase effective h_oe. It was designed for an era when “high performance” transistors weren’t the norm.

Matters a lot more in the VAS, where the ZTX devices are pretty nice.

 

[GlennW]

Your questions:
1. Version 4.5.
2. Rail voltage is +/-62. R13 and R14 are adjusted according to manual.
3. Fit of transistors with ECB pinout - you have to bend the leads a bit to fit the circuit board pattern. I used MPSA06 transistors for the bias circuit Q7, all other small transistors are 2SA992 and 2SC1845.
4. I used circuit boards from ebay seller amplifi_07. These generally follow the Leach design but are a bit smaller and include the Zobel components on the board. I wish there was more space for emitter resistors R45-R48, other than than I found the boards very nice.
5. Built about 3 years ago. The amplifier is in a shop area and only used occasionally.

Glenn

Glenn, could you elaborate on:
1) Was this the 4.5 version of the leach amp
2) what rail voltages did you use
3) how did you fit the KSC1845/992 in
4) did you make the boards or use boards from Mr Leach
5) how long ago did you build the amp and how long has it been in service?