freestompboxes.org

Hi,

I was having problems with hum from my pedals when I noticed that I didn't have the problem if I used the same power supply to power my Boss ME-X and powered the pedals from the ME-X power output.

I looked at the circuit in the ME-X Service Manual and saw that there is a voltage regulator circuit in the ME-X with additional filtering. So, I figured it would be a good project to build a copy of it!

However, some of the components are proving problematic to find. eg:

DSS310H (this is actually an EMI Suppressor)
S5688G diode
RD10MB2 zener diode
2SD20120 transistor

The rest of the parts are capacitors & and resistors so are easy to source.

Has anyone got any idea about sourcing the above parts, or equivalents?

Thanks,

R.

DSS310H...........https://cgi.ebay.com.sg/ws/eBayISAPI.dl ... 0596793520

S5688G..............1A, 400V PRV rectifying diode............1N4004

RD10MB2...........10V, 200mW zener diode...................1N5240B (0.5W)

2SD20120.........NO RECORD....Possible 2SD2012....NPN Power Transistor, 60V, 3A, hfe 100-320 in TO-220 pkg.

Thanks - that's a great help.

How did you find that ebay listing? When I search for DSS310H I get no results??

I eventually found most of the parts at Farnells, except the transistor, which I think it is indeed a 2SD2012 (looking at the rest of the circuit, the suffix is a letter 'O' rather than a digit '0'). However, Farnell don't stock it, and I can't find an equivalent. Grrr.

R.

I don't suppose you can suggest a suitable alternative for the 2SD2012? I can't find them anywhere.

Cheers,

R.

Without taking a look at the schematic the ratings of that transistor suggest it's being used as a voltage supply regulator. In that case all transistors that match the ratings Pokeypete wrote above will do the job without a hassle.

Dirk,

Is the hfe value important? I can find a few with 60V/3A but lower hfe.

I'll try and post the schematic.

R.

robinbowes wrote:I don't suppose you can suggest a suitable alternative for the 2SD2012? I can't find them anywhere.

Cheers,

R.

WES Components in Sydney, Australia stock 2SD2012 transistors at AUS$1.35 each.

They have a huge range of Asian transistors and IC's in stock.

Unfortunately, their website does not yet have a comprehensive catalogue online, so it's best to send specific queries by -

Email: sales@wes.net.au

Fax: +61 2 97166015

Web: http://www.wes.net.au

They send out a phone book sized print catalogue every 3 or 4 years, and fortnightly updates by email.

I found the 2SD2012 in the most recent edition.

robinbowes wrote:Dirk,

Is the hfe value important? I can find a few with 60V/3A but lower hfe.

I'll try and post the schematic.

R.

Just checked.
Not at all. The transistor is used as I expected and does not do any amplification.

Dirk,

That's great - thanks! However, would you believe that there doesn't appear to be anything suitable from Farnell that isn't "US Stock" (costs £15.95 to add to order). I *could* buy from elsewhere, but Farnell have all the other stuff (see below). Surely they have *some* sort of tranny that will do the job???

For the other parts, I'm going to use:

BL03RN2 - http://uk.farnell.com/jsp/search/produc ... sku=581124
DSS310H - http://uk.farnell.com/jsp/search/produc ... sku=580855
RD10MB2 - http://uk.farnell.com/jsp/search/produc ... ku=1467594
S5688G - http://uk.farnell.com/jsp/search/produc ... ku=9556109

and for completeness:

470uF/16v - http://uk.farnell.com/jsp/search/produc ... ku=8767157
120R - http://uk.farnell.com/jsp/search/produc ... ku=1612518
0.1uF - http://uk.farnell.com/jsp/search/produc ... ku=1469310

R.

Ah! the friggin US Stock.... drives me nuts more often than not....

A simple MJE18x will do the job for you

Thanks!

There's actually only two of those listed - damned US Stock again!

An MJE18004G for example?

And I suppose I'll need an heatsink, insulating kit and mounting kit?

R.

The Boss / Roland engineers may have chosen the 2SD2012-O transistor for this low-dropout discrete regulator circuit to take advantage of it's high current gain, unusual for a 25 watt / 3 Amp type.

The ground pin current of Q3 in a series low-dropout regulator is inversely proportional to the current gain, so using a 2SD2012 would keep the current through the Zener diode, resistors and capacitors in series with Q3's base low, while allowing good regulation at high load currents.

robinbowes wrote:Thanks!

There's actually only two of those listed - damned US Stock again!

An MJE18004G for example?

And I suppose I'll need an heatsink, insulating kit and mounting kit?

R.

Whoops... thouhght that was UK,

BD241C, BD242CG, BD243C, BD203 etc etc.


And bear in mind the trannie may be 3Amps but I wonder if the ME will draw that current

blackbunny wrote:The Boss / Roland engineers may have chosen the 2SD2012-O transistor for this low-dropout discrete regulator circuit to take advantage of it's high current gain, unusual for a 25 watt / 3 Amp type.

The ground pin current of Q3 in a series low-dropout regulator is inversely proportional to the current gain, so using a 2SD2012 would keep the current through the Zener diode, resistors and capacitors in series with Q3's base low, while allowing good regulation at high load currents.

Hmmm. If only I could find a UK source...

R.

Looking at the schematic, I see what I presume are two symbols for ground. What would be the difference between them? Is one signal ground and the other "earth" ?

R.

I see 3! There's a 3rd on the left. That's case ground that connects to the housing only.

The other 2 are connected to each other but, in the physical design, most likely at 1 point only. I suspect one being logic ground for all digital stuff and one the analog ground for all analog electronics.

And therefor of no concern for your application.

Dirk_Hendrik wrote:I see 3! There's a 3rd on the left. That's case ground that connects to the housing only.

The other 2 are connected to each other but, in the physical design, most likely at 1 point only. I suspect one being logic ground for all digital stuff and one the analog ground for all analog electronics.

And therefor of no concern for your application.

Yes, that looks right to me too. There is a Main / DC ground, a Digital ground that is coupled to DC ground via L20 (BL03RN2 inductor) and a case / AC ground that is coupled to DC ground via C42 (0.1 ceramic) and to Digital ground via C512 (0.1 ceramic).

Would it be a good idea to float the 9vDC + and - outputs and rely on the input jacks of the effects units being powered for ground reference? This seems to be the method used by some "big name" power supplies such as the Voodoo Labs Pedal Power units.

blackbunny wrote:

Dirk_Hendrik wrote:I see 3! There's a 3rd on the left. That's case ground that connects to the housing only.

The other 2 are connected to each other but, in the physical design, most likely at 1 point only. I suspect one being logic ground for all digital stuff and one the analog ground for all analog electronics.

And therefor of no concern for your application.

Yes, that looks right to me too. There is a Main / DC ground, a Digital ground that is coupled to DC ground via L20 (BL03RN2 inductor) and a case / AC ground that is coupled to DC ground via C42 (0.1 ceramic) and to Digital ground via C512 (0.1 ceramic).

Would it be a good idea to float the 9vDC + and - outputs and rely on the input jacks of the effects units being powered for ground reference? This seems to be the method used by some "big name" power supplies such as the Voodoo Labs Pedal Power units.

There will be no signal ground in this build - it's purely a power supply so I think what I'll do is use only the case ground, and leave +ve & -ve floating inside the case.

I still need to find a suitable source for the regulator BJT. I'm not sure if the alternatives suggested by Dirk will be OK after blackbunny's comments about the current gain. All the alternatives are relatively low hfe devices.

Also, I'm debating whether or not to design/build a PCB for this and make a few of them. When coupled with a cheap switching supply it would make a great alternative to the more expensive PSUs available.

Thoughts?

R.

robinbowes wrote:There will be no signal ground in this build - it's purely a power supply so I think what I'll do is use only the case ground, and leave +ve & -ve floating inside the case.

Yep, that should work fine.

robinbowes wrote:I still need to find a suitable source for the regulator BJT. I'm not sure if the alternatives suggested by Dirk will be OK after blackbunny's comments about the current gain. All the alternatives are relatively low hfe devices.

I am happy to contact WES Components on Monday and confirm that they are in stock. If so, I will add some 2SD2010's to my next WES parts order which I will fax to them on Tuesday morning (Australian Central Time). I am happy to post some to you at cost price, ie, what WES charge me + actual postage to you.

Alternatively, you could consider an easier-to-find BJT with good HFE but lesser current rating, if you don't need to draw 3 Amps from the regulator circuit.

For example, a BD139 (classification 10) has a HFE range of 63-160 and a BD139 (classification 16) has a HFE range of 100-250, with a maximum continuous Collector current of 1.5 Amps and PTOT of 12.5 watts.
The BD139 (16) would make a good precision low-dropout regulator with low base current similar to the 2SD2010 if you ensure that total emitter / load current never exceeds 1.5 Amps.

robinbowes wrote:Also, I'm debating whether or not to design/build a PCB for this and make a few of them. When coupled with a cheap switching supply it would make a great alternative to the more expensive PSUs available.

That sounds good to me. This is a very high quality discrete regulator design. The Boss ME-X discrete regulator can supply up to 3 Amps with a very high degree of regulation, so even a 1.5 Amp (BD139) version would be great.

Of course you could just build a simple IC regulator circuit like the VL Pedal Power and similar clones which use 78L09 (100mA) or 7809 (1A) IC regulators which will work well enough. I usually use LM317 adjustable regulators for better performance / lower noise with a maximum output current limit of 1.5 amps.

Or.....get the 2SD2010 BJT's and build superior quality 3 Amp regulated supplies!

Thanks very much for the kind offer, but I suspect the postage would be prohibitively expensive. Any idea how much it would cost for, say, 10 units?

I can actually find the odd one or two of these here in the UK (eg. on ebay. etc) but I was hoping for a more reliable source in the event I want to buy a qty.

My idea would be to build a regulator unit that takes any of the commonly available 9V switching supplies and applies regulation. Most of those are only around 1A-1.2A so the BD139(16) would prob. be suitable. However, no harm in over-engineering things

I have a couple of questions about the circuit...

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?

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?

Thanks,

R.