dual API 312 work in progress / pics

[martthie_08]

I have gotten a little stuck on this Fabio 312 project that I've been working on. Although there is already tons of information out on the 312, I thought it would be nice to have "another" thread, I'm planning to test some different opamps, starting with a diy melcor 1731 guinness: to Volker), then on to Fabio's 25.20 guinness: to Fabio), diy JE-990 guinness: to Volker and Kubi) and there might be interesting things coming up. The preamp board itself has Cinemag CMMI-8 input and Profile 4804 output transformers guinness: to Peter), the line input, servo and unbalanced output options were omitted. The relays are all 12V and I have (been trying to) adapt the psu for this.



Now the first problem: The PSU (Fabio)
I am using a 24V-0-24V mains transfomer which yields about 30V AC no load and 40V DC after rectification. At the moment I have four 2200uF / 35V caps with R1 / R2 = 10R between the two respectively. I didn't pay attention to this initially, but of course the DC voltages are a tad too high for the caps. Can I replace all four caps with 1000uF caps with the idea, that the pos. and neg. rails will be powering only two opamps? I have 1000uF / 63V caps at hand that would fit the pcb perfectly.

Next thing is that I am feeding the input of the 12V relay power regulator (7812 for 12V) the unregulated DC coming off of the R1 / C1 junction with a 470R (5 watter, under the board). This works fine, I checked the output voltage of the regulator (11,76V), but as soon as I start engaging the relays, the voltage drops by approx. 1V per engaged relay. How can I trace this problem down?



Regarding the grounding of the PSU, what is the best way to do this. At the moment, the board is not grounded at all, do all four voltage grounds (+16V / -16V / 12V / 48V) need a seperate path to star ground near the IEC socket? Sorry if this had been discussed before, but I couldn't find anything and I'll promise to make a webpage that will organise this information, so that this stuff doesn't have to be answered every three months :grin:

::::::::::: edit: added some more info's ::::::::::
That's all for now, thank you!
cheers, Marten

 

[Handcrafted Tone]

you now have 312 posts.... :grin:

 

[martthie_08]

Rob, I can't wait to get to my 1073th and 1081th post!!! :grin: :thumb:

 

[AnalogPackrat]

Marten,

Looks nice so far. On your 12V reg question. I think your dropping resistor (470R) is too big. You've got 40V at one end and a bit over 12V at the other assuming no load:

I = V/R = (40V-15V) / 470R = 53mA

I've built a pair of these myself (also with 12V relays, but using an 18-0-18 power toroid), but I can't recall the current demand of the relay coils. I want to say it was around 100mA each. Anyway, what seems to happening is that you're dropping too much voltage before the regulator and it can't regulate.

You want to determine the maximum current draw (including all relays and any LED indicators you might have hanging off the 12V rail) and then back-figure that dropping R so that you always have about 15-16V at the input to the regulator. Then you can figure the worst case power dissipation in the resistor and get one that can take the heat. On mine I just used a jumper and a bigger heatsink on the 12V regulator, but I didn't have quite as much voltage after rectification.

Good luck,
A P

 

[martthie_08]

Thanks AP, you may be right, the 470R was just a value I picked up on one of the threads, the maximum current draw for the 12V part will be the power on led plus three relays per channel + three led's per channel:
1x 10mA
3x 140mA
3x 140mA
3x 10mA
3x 10mA

sums up to 910mA (wow!), so to get the voltage down by 25V this would result in a resistor R = voltage / current = 25V / 0.91A = 27.5 R, am I getting this right?

 

[AnalogPackrat]

And do the math for power. I just made a quick and dirty spreadsheet to try different R options. No matter what you do, with 910mA of current draw on your 12V rail you'll be dissipating a total of about 20W of waste heat in the dropping R and the regulator. You get to choose how much in each part, but 20W is a lot.

BTW, your voltage at R1-C1 node should be more like 35V under normal operating conditions with that transformer. If it were me, I would:

1) use a 15-0-15 or 18-0-18 toroid
2) use high brightness LEDs and run them at 3-4mA instead of 10mA

This will get you into more reasonable territory. Save that big toroid for another project...

A P

 

[martthie_08]

yeah, I should have used a 18-0-18v trafo from the start, but at the time I started the project I was looking into 24V opamps. so I went for a dual 24V.
The diodes don't pull an awful lot compared to the relay current requirements, so I'll leave 'em as they are.

Maybe I'll try a 10R / 11 watt resistor before I order another trafo, this should knock the voltage down a bit and put less strain on the 12V regulator. If I built another one of these preamps I'd definately use 24V relays and a 7824 regulator to go with that.

keep wondering about the psu grounding scheme - anyone?

 

[3nity]

Marten, did you tried the new transformer??
I used 9 volts relays. So i use a 7809. i Had a few problems with the switches engaged produced a oscilation into the preamp.

the transformer i used it is a 22V-0v-22V i got laying around.
Do you think i'm better change the transformer too??
I'm still figuring out what may cause it.
Thank you.

 

[3nity]

Marten, did you tried the new transformer??
I used 9 volts relays. So i use a 7809. i Had a few problems with the switches engaged produced a oscilation into the preamp.

the transformer i used it is a 22V-0v-22V i got laying around.
Do you think i'm better change the transformer too??
I'm still figuring out what may cause it.
Thank you.

 

[martthie_08]

Hello, I didn't find time to work on the project the last couple of days. I'll report back for sure when I get it going.