Philicorda 751 (tubes) Weird Circuitry

[sarakisof]

Hello everyone, after restored (months ago) was messing around and trying to record my Philicorda GM 751 the other day, noticed that smthng weird was happening with the reverb. Then i soon realised (and remembered from schemo) that in fact that's the way it is designed somehow.

When reverb is off everything is normal panned.
By switching reverb on, one speaker is like dry and the other is wet mix reverb. By turning the rvrb control pot down the reverb speaker decreases till almost silence when pot is all the way down, so you hear the dry signal from the first speaker only.

Is this normal according to scheme? I'm lost a bit here. I mean it's a bit weird, see the case: let's say as you pressing a key and by having the rvrb mix pot all the way down you just turn the Reverb switch On and Off. Normally nothing should change, you should keep hearing dry signal in stereo. Now by switching to On you only hear music (dry) by one speaker!

https://www.sm5cbw.se/audio/philips/GM751.pdf

https://www.diyaudio.com/forums/instruments-and-amps/339218-philicorda-22gm751-reverb-missing.html


Plus, so to proper record it, i should use a diy 5 pin din to two mono jacks?? Cause in my all other Philicordas (transistor models wih "normal reverb" 🤣) i just used a 5pin din to just one mono jack by soldering positive and negative together, with no probs.
Say that because if i have to use two mono's for just line recording, then it will be strange again if i want also record in the same time its speakers, i will need three inputs then!🙄

[11-90-an]

Now I don't read german, so I google translated it, and this is what i get: 



From the block diagram, when the reverb switch is off it shows that the signal is split into 2 speakers, each split goes into pre-amp, a "control level" (dunno what that is, maybe a volume control?) and a power amplifier. After that, it goes to the speaker.

When the reverb is on, the signal goes through pre-amp, "control level" and power amp, then splits into 2: one goes to a speaker, one goes to the reverb tank, another amp, the reverb control pot, goes to another pre-amp, "control level", and a power amplifier to the other speaker. And that's why one speaker plays dry and the other plays wet.

You probably can just short the wet speaker and dry speaker outputs together with say, a 100Ω resistor if you want one mono out only

[Rob Strand]

The root of the problem is they are using the speaker amplifier as the reverb drive amplifier.    If you feed the reverb output back to the input of the same amplifier you can get internal feedback, like you get with a mic.   In this scenario you are forced to keep the reverb level low in order to keep below the positive feedback threshold.    The result of that is the reverb often sounds weak.

IIRC some early guitar amps did the same thing, and the complaint was the reverb was weak.   So in later revisions of the amp they changed the reverb to have it's own drive amp.   Can't remember what amp that was.

Anyway in order to avoid the above they have gone for a different solution.  The reverb is still driven by the speaker amp but the output doesn't feedback on itself, it goes to the other channel; you don't have this luxury on the guitar amp.   So you can crank the reverb up without the risk of positive feedback.   A kind of advantage to that is the reverb might sound a bit more spacious as the signals combined acoustically.

To record the signal with reverb you would need to combine the dry and reverb signal.  You could do it with a resistive mixer/divider with some judicious choice of the reverb level so it sounds like the speakers on the unit.

[sarakisof]

Yes exactly it now makes sense indeed.
Organ's line pre out is this i have circled in orange.
So i guess i should be ok if i used two mono's right?
And then another for micing it. Or like 11-90-an suggested above using a res.

Yeap Rob, the sound now is too "spacy" and I don't know if i like it like that. And the "prob" is you tutn down the pot a bit and you loose the Right speaker.
How could i "blend" them both speakers by having the opportunity to turn up/down the pot taste wisely?
Rob you probably answered this, but my apologies I don't get it  don't forget English is not my native language, i am Greek  γειά σου 

.You could do it with a resistive mixer/divider with some judicious choice of the reverb level so it sounds like the speakers on the unit.   

[sarakisof]

Oh god, i could also do this in DAW right? Organ two mono outs ---> two channels in daw dry n wet signal. And then mix the accordingly ---> sum up channels to one stereo 

Or no?

[Rob Strand]

Make a cable like this,



R1 220k
R2 220k
R3 47k

U1 to DIN pin 5
U2 to DIN pin 3
gnd to DIN pin 2

Take output from:  OUT and ground.

R1 and R2 high enough value to prevents loading.
R3 trade-off between signal level and amount of reduction of reverb signal feeding back into DIN pin 5

[sarakisof]

Ok i see, stereo to mono circuit sub. Thanks Rob.

I will first build a small DIN to female 6.3 jack adapter, and then i will make two cables and use accordingly.
The Rob's one if i want one mono, and another "without circuit", just a simple Y cable from pins 2,3,5 DIN to two 6.3mm mono jacks.

Will be no problems with the second choice right? I mean that's the way it was "designed to work".

[Rob Strand]

Will be no problems with the second choice right? I mean that's the way it was "designed to work".

The second one will drop the level.   

There's two trade-offs:

- One trade-off is not loading down the existing signal.   That's a achieved by using the high value of 220k.  The original circuit didn't seem to care too much about external things loading it down!

- The second trade-off is preventing the reverb signal feedback into the non-reverb channel.   The potential problem signal path is:

Reverb signal on DIN pin 3 -->   R2 -->  output terminal ---> Signal back into R1 ---> then reverb back into DIN pin 5.


I've given quite safe/cautious values.   They should work without any problems just that there is a noticeable signal drop.

If the signal drop is too much you could gain some signal level with some slightly less cautious values.   Raise R3 to say 100k.    Then say reduce the 220k's to R1=100k and R2 = 100k.

The idea of the divider is fine is just a matter of finding a nice balance with the trade-offs.

You could do "better" version by modding the unit but it's best to leave the unit unmodified and make do with the trade-offs.