Welcome to Solid State Guitar Amp Forum | DIY Guitar Amplifiers. Please login or sign up.

March 28, 2024, 10:25:08 PM

Login with username, password and session length

Recent Posts

 

acoustic 260 relay replacement

Started by ilyaa, July 18, 2016, 06:54:17 PM

Previous topic - Next topic

ilyaa

bringing this one back from the dead

looks like the reed relay died whenever and someone had rewired the bright switch to bridge the gap. unfortunately, as a result, the fuzz does not work - cannot be switched into the signal.

ive tried snooping around online for a replacement relay but really no dice!

check out page 23 of the schematic

http://bmamps.com/Schematics/acoustic/Acoustic_260_Service_Manual.pdf

part number GB 651C 10?

cant find it - not sure what a good way to find a replacement would be - any advice appreciated thanks!!!


galaxiex

If it ain't broke I'll fix it until it is.

J M Fahey

Mouser *should*  have a compatible one.

Enzo

Is voltage getting to the relay coil?   As I read the signal, the relay works off +25 through R43 and S5. (S5 is part of gain control R41)  So S5 has to be closed, does it close?  Do you have +25v on R43?  And with the switch closed does that +25v reach the relay itself?

Does the coil read open?

When you try to switch, does ANYTHING change?

I have an SPDT reed relay, about 480 ohm coil, I'd part with.

But really, they used a reed, but most any small signal relay will work.  If you are not hung up on fitting the foot print of the original any format small relay can be mounted dead-bug style (Glued on upside down, legs in the air, short wires down to the board)

I don't know what voltage the original part was rated at, but you can use other voltages by changing R43.

I suspect R43 is as big and heavy as it is simply to serve the footswitch.  Without a footswitch, a 24v relay would work,  but the footswitch turns it off by grounding the supply to the relay coil, and that resistor limits the current under those condition.  SO I'd guess a 5v relay was used, but that is a guess.

ilyaa

i guess i wasnt 100% clear earlier: there is no relay - before this amp got to me, it was removed and someone actually rewired the bright switch to bridge the gap that the relay was supposed to bridge. but, like i said, as a result there's no way to turn the fuzz on. and on top of that the bright switch doesn't work.

so my challenge was finding a relay to replace the existing one but i wasn't quite what a suitable replacement would be.

enzo, how are you calculating what voltage the coil wants to see? just assuming a lowish coil resistance and voltage divider with R43 from there?

i'd be totally fine with dead-bug style - just wanted to make sure i'd have the right voltage relay.

Enzo

R43 is largish and 2 watts.  That would be overkill for just a relay coil, and also considering it runs on a 25v rail and they make 24v relays commonly, I have to look and see that the only reason for that resistor is to allow a footswitch to ground it.

Relays tend to come in 5/6v, 12v, 24v, 48v increments.

At 8200 ohms, even a dead short - as in the footswitch is on - only draws 3ma through it, so I figure the relay itself cannot need very much voltage.  So I assumed for discussion a 5v relay was likely.  Something that will close on low current drive.

The relay itself is SPDT, so I am not sure just where the gap is your switch has crossed.  As drawn it is in clean channel mode, though the input signal is fed to the extra gain stages for dirt all the time.  The relay then would switch over to use the output of the dirt stages to feed R7A and Q2.

You could tack in a plain old SPDT toggle switch to those three terminals to verify the circuit works or not.   Feeding a regular input ought to put signal on the output of Q1 at C4A.  Whether you can hear it or not, you should have distorted signal coming out of Q10 and C30, just to make sure it works.  Then the relay selected either the original signal or the C30 signal to feed into R7A.  So you can inject a signal at that point to verify the rest of the amp works after that point.  If all those sections work, and the switch in place of relay functions, then  all we need do is find a low current relay to stuff in there.

The board photo looks like there is plenty of room for a relay, so should be no problem to dead bug one in place.   Looking at Mouser, this looks like maybe a candidate:
http://www.mouser.com/ds/2/262/Relay_NP_Series_E-2806.pdf

SPDT is called form C, a make/break switch.  The 5v version will close at 3.5v and hold down to 0.75v.  That huge resistor is making it tough for us.

So far i only looked at reeds, and I suspect they will be the most sensitive coil wise.  But there are some pretty low current regular relays there too.   And I imagine you could alter that R43 to make things work a little better.

At least that is how I see it.

ilyaa

yay!

had a suitable 12v relay around. popped a 10K resistor across the 8.2k so i could get the right voltage and we're back in business! fuzz is working and bright switch is working.

two last issues:

1) the first main issue in the amp was that the LDR/lamp had failed in the tremolo section and was causing output problems. it was one of these: http://www.ebay.com/itm/Sigma-301R1-24-Datacell-Optical-Relay-24-v-lamp-1-LDR-5-Pieces-/281251030611. troubleshot it, determined that the LDR was bad as were Q18/17 darlington pair (the LFO signal as not getting to the neon). replaced those, got the neon to flash. then put it an LDR i had around and wrapped it all up together with the neon. tremolo works BUT the depth/intensity is not quite as rad as it could be. tried funking with the bias but no dice. figure that sigma datacel probably rules when it works right. worth it to get another one?

2) the output transistors are getting pretty damn hot - as is their heatsink and the chassis with it. ive ran it for about 20 minutes at nearish to full power and no problems but the heat.....should i be concerned?

Enzo

You are calling it a neon, did you use an actual neon lamp in there instead of a 24v incandescent?

What part of the original was bad, the bulb or the sensor? It LOOKS like the original part can be taken apart by unbending the tabs, so you could test the resistance of the photocell by measuring it with light on it and a gain when dark.  That will tell you if the cell works.  A small bulb could be installed into it in place of a bad bulb.

Power amp.

Does it get hot just idling?  Normal for an amp to get hot when cranked for extended period.  Set the bias.  Monitor mains current draw, turn the bias to coldest setting, watching mains current.  Now turn control slowly upwards until you first see mains current start to rise with the adjustment. Stop and back off a hair.  That should be coolest setting and the point where crossover distortion goes away.

ilyaa

#8
whoops ya dono why i said neon. incandescent bulb.

yeah the original can be taken apart - i did that and the photocell was bad, bulb OK.

what i was saying was that my homemade rig is probably not as sensitive/effective as a real one - as far as photocell speed and correct light/dark resistances, etc.....

it DOES get hot just idling. fiddled with the bias and i think we're doing much better now. as it was before, it was idling at like 1.5A mains draw. got it down to 0.3A - much nicer, just gets a little warm idling. and at full power its at about 1.5A. sound about right?

on another interesting note: whoever worked on this amp before did a 'quick-fix' that's new to me! the mains fuse was blown, but instead of replacing it, they wrapped it in tin foil! so it was just an unfused straight short!!! wow what a dangerous corner to cut!



Enzo

Anyone in the repair business sees foil fuses, but it is still scary whenever we see it.

Random thoughts:
Is C41 leaky?  ANy DC on the photocell side???
Unlike a lot of systems like Fender where the oscillator draws current through the bulb, in this the bulb is in series with a resistor to +25, but the controlling transistor is parallel the bulb.So it would seem to me that the bulb part is less sensitive to the circuit, meaning we could maybe have an LED work there.  May have to add some resistance to something.  And the point of that would be we might get away with an LED based optocoupler like a VActrol in place of this.
You might just need to use a diffrent photocell.  Jameco has a good selection.  and they list light and dark resistance for each:
http://www.jameco.com/shop/StoreCatalogDrillDownView?langId=-1&storeId=10001&catalogId=10001&freeText=photocell&search_type=jamecoall

I have bought photocells from them for years.

ilyaa

yeah c41 was a bit leaky i had already replaced it -

as far as choosing a photocell that's good for this kind of application, what are the useful criteria? large dark/light disparity?

Enzo

WHy not characterize what you have, then we can compare it to the specs of others available.  We may find why it has limited range.  And remember, the specs are from dark to light, but we don;t know if your bulb is as bright at peak as their light side.  Dark is dark...hopefully.  I tend to prefer ones with larger range but also a relatively low on resistance.   In the high resistance end, the difference between 1 meg and 20 meg is not likely all that much.  But the differnce btween 2k and 200k for the low end might make all the difference in the world.  So to me it is a judgement call looking at the particular schematic.

Your bias control probably allows you to get the low end completely dark and maybe the high end completely bright, but you may not have access to the whole range of that in the operating circuit.  Ever change the pot in a wah wah?  The gear thing only turns the pot part way, you have to set (bias) the pot mechanically so the motion you have available best fits the resistance range.  Bias on this trem circuit is like that, it moves the range of change.

ilyaa

tested the LDR i was using - it was going from about 1K to 2M. found another one i had around that went from about 200 ohms to 4M. better range.

on top of that, i replaced the incandescent bulb, which wasn't too bright, with a yellow LED. didn't change anything in the circuit - seems to work fine.

with the new LDR and the LED the tremolo sounds much better! deeper and with a more discernible range.

so far so good - thanks, enzo!