Obscure Japanese semi-conductor equivalents

blackbunny · Post by **blackbunny** » 05 Mar 2011, 15:55

Here's a link to BD135 / 16 BJT's at Musikding - scroll down toward the bottom of the page.

They are a lower voltage rated BD139 type with the same high HFE and 1.5A current rating.
Should be good for a 1.5A ME-X type regulator build.

http://www.musikding.de/index.php/cat/c ... ersal.html

robinbowes wrote:I may have a need for an 18v version (would obviously need an 18v input). How is the output voltage set? Obviously something to do with the 10v zener... but what?

The 10 volt zener does indeed set the output voltage: 10v - 0.6 volts drop across Q3 = 9.4v regulated, +- around 0.2v for zener tolerances.

For 18 volts output, you could use a 19v zener (??) or stack a 9v1 zener in series with a 10v zener for about 18.5v regulated output.

Make sure you use 25v or 35v electro capacitors!

You can find precision zeners and zeners in odd values, but they may be hard to get, so best to stick to standard values. Most effects circuits (excluding Ge fuzzfaces) operate perfectly if you supply clean, well-filtered DC at their rated voltage +- 1.5v or so.

robinbowes wrote:Is the resistor network across the output needed? Or is that just there as a voltage divider to obtain 4.5V for the input to IC1B?

Not needed. Just there to derive 4.5v (bias) supply to op-amps inside the ME-X as you correctly state.

robinbowes · Post by **robinbowes** » 05 Mar 2011, 15:58

Hmm, alternatively, the LM338T will regulate up to 5A and is only £1.14/unit if bought in quantities of 10 or over, ie. about the same cost as the transistor alone in the discrete circuit!

Perhaps I'm over-complicating things...

R.

robinbowes · Post by **robinbowes** » 05 Mar 2011, 16:07

Sorry, posted last reply before I saw your reply...

Thanks very much for the explanation. My first thought was 10 - 0.6 = 9.4v but I guess I was looking for 9.0! It's been a long time since my last electronics exam (21 years, if memory serves!)

I move house in 7 days, so I think I'll mull it over for now and re-visit when I get my electronics stuff unpacked in the new house!

Thanks again for all the help + information. Much appreciated.

R.

blackbunny · Post by **blackbunny** » 05 Mar 2011, 16:14

robinbowes wrote:Hmm, alternatively, the LM338T will regulate up to 5A and is only £1.14/unit if bought in quantities of 10 or over, ie. about the same cost as the transistor alone in the discrete circuit!

Perhaps I'm over-complicating things...

R.

I know that feeling. I have been perfectly happy with my home-made LM317T regulator circuits for pedalboards for about 20 years, and have had excellent results using the LM338T for low noise DC heater supplies in valve preamps.

I would say that the law of diminishing returns kicks in about now: why put in 300% more effort and 300% higher parts costs for 10% better regulation?

All the same, hats off to the Roland (over-) engineers for an exemplary design in the ME-X.

blackbunny · Post by **blackbunny** » 05 Mar 2011, 16:25

robinbowes wrote:Sorry, posted last reply before I saw your reply...

Likewise...

robinbowes wrote:Thanks again for all the help + information. Much appreciated.

No worries. It certainly got whatever brain cells are left inside my fuzzy dome earning their keep!

Cheers,

BB

Hides-His-Eyes · Post by **Hides-His-Eyes** » 05 Mar 2011, 17:01

Anything you can put in an envelope is really cheap to post from the US; it's whole pedals etc. that get silly expensive.

blackbunny · Post by **blackbunny** » 06 Mar 2011, 05:56

Hides-His-Eyes wrote:Anything you can put in an envelope is really cheap to post from the US; it's whole pedals etc. that get silly expensive.

Not quite so cheap from Australia to the UK. Australia Post website international rates calculator yields a cost of AU$13.50 to send a C5 sized envelope weighing between 125 and 250 grams by Express Post. Allowing for a C5 size padded envelope, you might fit 15-20 x 2SD2012 transistors in there - T220-type cases are fairly heavy.

Regular post would go by sea and could take 8 to 12 weeks, at about half the cost of Express Post.

I'm tempted to build a couple of the MX-E style 9v regulated supplies myself, but with my current workload & backlog it would take a few weeks to get around to it. I will order a few 2SD2012's anyway, just in case.

robinbowes · Post by **robinbowes** » 25 Mar 2011, 22:59

Just following up on this...

I've successfully moved house and have placed an order with Mouser for a load of bits to build the Boss regulator circuit, along with a few bulk purchases from ebay. I also got some LM317Ts and L7809CV regulators so I can compare their performance to the ME-X discrete design.

I plan to build myself a 9V prototype, and also build my mate an 18V version for his Lone Wolf Harp Attack pedal. Then I'll see about getting a small circuit board built and maybe make a few to sell on to recoup some costs!

R.

DeanM · Post by **DeanM** » 09 Apr 2011, 23:12

robinbowes wrote:I also got some LM317Ts and L7809CV regulators so I can compare their performance to the ME-X discrete design.

That would be interesting to hear about the results.

I've just recieved postage yesterday of 50x 78L09 TO-92 9V regulators and plan to build power supplies with them. Was pretty price, just over €6 for the 50!

robinbowes · Post by **robinbowes** » 09 Jan 2012, 09:58

Can you believe I've only just got round to building one of these?

I built one on a small piece of stripboard and housed it in a 1590A enclosure (with a power socket in each end). PCB mount sockets would be good as they would also double as PCB mounts. I may build one into a larger box for my pedal board, and also include input/output/FX send/return etc.

So, how does it perform? Well, I've built an 18v version which I have no way of testing as I have no 18v pedals - doh! It seems to measure OK though - on the 'scope, the output looks a lot cleaner than the input.

I shall report back when I've tested with a real pedal.

R.

robinbowes · Post by **robinbowes** » 09 Jan 2012, 21:35

I wonder...

Could I produce multiple, isolated outputs from a single input by simply using multiple transistors? Hmm, I suppose they would not be truly isolated - they would share a common ground / -ve connection?

R.

blackbunny · Post by **blackbunny** » 10 Jan 2012, 13:51

robinbowes wrote:Could I produce multiple, isolated outputs from a single input by simply using multiple transistors? Hmm, I suppose they would not be truly isolated - they would share a common ground / -ve connection?

You can produce multiple isolated DC outputs from a single AC tap by using several bridge rectifiers, each followed by a filter cap, regulator and output capacitor. Each DC output can then float if you don't join the negative outputs together.

This topic discusses some approaches to isolated DC outputs: https://www.freestompboxes.org/viewtopic ... =9v+supply

robinbowes · Post by **robinbowes** » 03 Apr 2012, 22:00

So, I tested the 18v version of this circuit with the pedal it was built for (a Lone Wolf Harp Attack).

It didn't seem to like it - the output voltage appeared to be too low. With 2 x 9V batteries in, the LED glowed much more brightly than when powered from the 18V DC supply. I tried a higher-rated zener (20v instead of 19.1, I think) but it didn't seem to make much difference.

I wonder, is there some inherent limitation in the circuit/components used that would limit the output voltage?

This is the schematic, btw: https://www.freestompboxes.org/download/ ... p?id=11682

R.