MBSID design help appreciated

[SounDuke]

Hello everyone, long time no see ^_^

I'm planning my quad MBSID8580 and I'd like to obtain precious suggestions from who has more experience than me.

I'd like to build the Step C, 4 cores and 4 Sid8580 but I'm planning to add some personal features:

-1 Oscilloscope

-2 Oled 2x20 display

-3 Lead battery powered

N.1: I've built a little oscilloscope using a pic 16f877. It is stuffed with a KS0107 graphic display with backlight. It draws about 100mA.

N.2: I'm equipping the MB with an oled display to reduce power consumption.

N.3: I was thinking to power the whole synth and integrated oscilloscope with a Lead battery 12V 7Ah.

What do you think? Is it feasible? I'm concerned about the regulator, as I think I'll go beyond the 1,5 Ah.

I'd like to regulate the 12V through a 7809 and supply the 9V to the sids, then connect this 9V line to the 1st core 7805 to supply both the oscilloscope and the 4 cores. What do you think?

[SLP]

Why do you want to use batteries ??

It could become a bit heavy...

can you give me the schematic and the code for the scope?

thx,

matthias

[SounDuke]

sorry, I did that scope lots of time ago, it was on an issue of EPE magazine :(

The battery is because I'm always traveling, and for the steadiness of current too, I would like to know if the separated 9V and 5V is absolutely needed or if it is possible to supply the whole MB with one 12V line. Thanks

[TK.]

Maybe it makes sense to use a special low-drop regulator for such a project. There should be good schematics available in the web, just google for "low-drop 5v schematic"

Best Regards, Thorsten.

[SounDuke]

Thanks Thorsten, if I'm not wrong I argue that the low dropout option is to avoid malfunctioning derived by the discharge of the battery and the consequent lowering of voltage. Since I'm not sure wether powering it with a psu or the battery, I was wondering what is the best way out.

regarding the PSU solution:

For my MBLC I used 2 toroids which accept various input currents (110-220V 50/60Hz), because I travel to different places in the world and I fear the problems which could occur for example by plugging normal PSU's I could buy with Tokyo's current network (100V 50Hz) which is different by the European one. I own a C128 psu which accepts 220V/50Hz, so I can't use it without a stepper-down converter.  Maybe the best configuration for the MBSID is to design a 2 toroid power supply like I did on my MBLC.

regarding the Lead battery solution:

If I use a low dropout 5V schematic, how to obtain the 9V line? is it ok to power the 4 Sids regulating directly the 12V with the 7809 on the Sid module?

Thank you very much ^_^

[Lall]

Hi,

For the LDO on the 5V line, I'm not sure it's really useful. The battery or 7809 should have enough voltage for the usual ~2V drop anyway and it will not bring anything from power dissipation point of view. Maybe it would be more interesting for the 9V line as you're closer from the battery voltage.

Concerning the usual regulators, you have to watch out that the dissipated power is given by (Vout - Vin) * Iload. So for a load of 800mA on the 5V output with 12V input, you get 5.6 Watts to dissipate. In the end you have to foresee significant heatsink if you don't want to have a 7805 toaster ;). With the usual 220V approach, the good thing is that you can of course choose your transformer to minimize the voltage difference and reduce the power dissipated which is not possible with batteries.

I think that you have to look for the amount of current taken by each and every part and check what the best option is. Maybe you'll have to put several 78xx in parallel to have enough capacity. Another solution could also be to use a 78xx doing the regulation and another transistor to handle the actual big current, the schematic of this was posted already once here.

In any case, the power dissipated will be the same and the heatsink will have to be scaled accordingly. I was myself pretty surprised to see the potential load of the SEQ I'm working on right now though most is coming from the LCD's backlight LED.

As far as I know but I'm no expert in the field, the only workaround from power dissipation point of view is to use a switching power supply but that's usually not recommended for audio stuff.

Best regards,

Lall

[SounDuke]

could anybody tell me what's the overall power consumption of the 4 cores with a standard 20x2 backlight lcd and the sid modules logic, and the consumption of the 4 sid chips? I can calculate the power I need then adding the oscilloscope power consumption.

Thanks

[Lall]

As far as I remember from reading this forum, someone once said a Core was consuming around 100-120 mA.

The consumption of the DIN and DOUT alone is close to neglictable I believe but of course you need to count the consumption of the LEDs which you can derive from the resistors you've chosen. Concerning the SID, I must say I've no clue and I don't remember if it was part of the post mentioned above.

For the screen, you should take a look at the datasheet of the model you have and watch out if you have a LED or an EL backlight. Before looking at the projected power consumption of my SEQ, I had no clue what the difference was and there's a factor or more than 10 in power consumption. Anyway, the 20x2 LED backlight screen I own a consumption of 250-300mA.

For the SEQ ones i.e. 40x2, it's even worse as it goes up to 550-600mA which is amolst half of the overall power consumption.

Best regards,

Lall

[SounDuke]

Hi,

Thank you Lall, the power consumption of the oled display is something around 100mA, the problem is that the graphic display consumes something around 250mA with 3V. I think I should make it with a toroidal transformer with 2 secondaries 9V, each 1,67A. The question is: I have yet the 9V regulated by the toroidal, should I apply something before the sid modules? I argue that I can't directly plug one 9V secondary to the sid modules.

Thank you

[Lall]

Hi,

Indeed, you have to have the usual bridge rectifier to transform the AC into unregulated DC followed by the regulator between your transformer and the SID's 9V input.

The 9V AC output of your transformer should give you 12.6V peak voltage (Vpeak = Vrms * sqrt(2) or 9 * 1.4 in this case).

With the usual full wave recitifier, you'll loose two times the drop voltage of the diodes used in the bridge. This makes  ~0.6V * 2 leaving ~11.6V peak. You'll loose again still some voltage due to the capacitor used before the regulator, the bigger the capacitor the better.

In the end, you will have to use a LDO regulator I believe otherwise with a normal regulator you won't have enough voltage at its input to have it work properly.

Maybe you'll want to double check these computations/assumptions and also pay attention to the fact that the voltage at the output of a transformer has a tendency to go down as the current grows. Maybe you'll have to reserve one of the secondaries for the 9V stuff only to make sure that the current is limited and the drop at the transformer too.

Best regards,

Lall

[SounDuke]

Hi lall, thank you for the wise analysis. I bought for my MBLC a toroidal transformer with 2 secondaries, each 12V. To power the MF module I was able to put the usual Bridge rectifier ---- 2200mF--- 7812 ---- 100nF pattern, and the regulator had the required additional voltage to start regulating the current. So maybe I'll be able to reach the required voltage with the 9V secondaries.

in any case, I believe I can use one 9V secondary to power the sids and one 9V secondary to power the 5V logic/leds stuff. I'll put 2 7805 in parallel to obtain the 2A current I need for this, but for the sids I think you're right and I'll not be able to regulate to steady 9V the ac current of the secondary. So, can you suggest me a schematic for a suitable LDO?  Thank you

[Lall]

Hi,

As far as I've heard from colleagues using LDO frequently, good choices for standard LDO's can be found at TI or National. More specific and specialized ones can be found by Linear Technologies for example.

You have to use their parametric search tool to find the ones that best suit your needs and then compare the found items with what you can get from your shop/supplier/...

Looking at the TI website, I've found this one: http://focus.ti.com/lit/ds/symlink/tlv1117.pdf that could do the job apparently. There are also quite others to be found. In that document they explain shortly how to use it and the recommended conditions.

My colleague also brought my attention on the fact that the main voltage can vary quite a big +/-10% so does the secondary voltage. As a consequence, you can have quite some variations at the input of the regulator and therefore you need to choose it to support enough at the input for example 15V max.

Best regards,

Lall

[SounDuke]

Thank you lall. The LDO you proposed has 800 mA on output, isn't it too low?

[Lall]

Unfortunately I don't know.

I did look at the datasheets few months ago to see what the theoretical max load was and I think it was quite smaller than 800mA if I remember correctly.

Then I wanted to measure the load of the LCD, the core and the SID on my Step A SID but I've used an old PSU I made but never put in a box and the regulator moved I don't know how but finally I think I've got Vin instead of Vout on the 5V line. By chance I did my first experiment without the SID on-board but the PIC and the LCD are dead  :(  and I haven't repaired the stuff yet...

Maybe you can find that info on this forum, there was once a discussion about the load of the Core and other different parts so maybe the 9V line is also there.

Best regards,

Lall

[SounDuke]

I'm sorry for the incident. Anyway thank you for the infos