Altec 436C build COMPLETED! with questions

[college101]

Ok..I am back! Its been a while since my last post. Im finishing up my last class for my associates degree in Electrical Engineering. Right now I am in an Advanced AC class, and im having a blast in it.  I also recently passed the CETa exam....

So for my class project, I decided to build a P2P turret board 436C. Hammond PT,-Sowter input and output, some RCA tubes- did the darn thing right. I got an "A" on the project and my teacher is pretty impressed with it.

I have completed my burn in period, and want to make sure I have it biased correctly. I followed the original spec's in the widely passed around 436C manual online. There were only two small changes I made to the circuit, which I dont really count as "mods" -Which some people do for some other reason than I know.

1) I installed a 100 ohm adjustable pot in R8's place to help ensure I have the correct voltage across the honeywell VU meter I installed.**I know , i know- I couldn't get to a real Altec meter and had to pull the trigger with this one...

2) I replaced R13 Power Supply resistor with a 3 watt over a 1 watt... Its the little things right guys?

So, now I was expecting to be able to tune my voltage across the meter and get it to sit at idle, with the meter reading at zero, but right now,its around minus 20

BUT, if I play with the controls, varying voltage and input,threshold, and release, I can move the needle around...but this is with no input signal and the compressor at idle....

  Some helpful responses could get me going in the right direction! Thanks  College101

 

[zebra50]

Any chance your meter is wired in reverse? That would make your needle sit down to the left and move right with compression. You want it at zero and moving to the left.

I don't recall this being a problem, although I built mine about 8 years ago now! I used a sifam meter and did the same thing, fitted a pot to set zero. I think I used a bigger pot, maybe 1k.

I know some people think that these suck, but I think this was one of the best things I ever build, and still gets used on every session.

 

[college101]

I thought that at first zebra50-BUT- I re-checked my wiring, and the honeywell vu meter does have its connection points for signal labeled..  I followed that assuming it was right...Geuss it could be wrong, I will try and check it out later..

I thought just from a biasing standpoint, that I would have adjust that resistor to zero the vu meter

Right now playing with the controls on the faceplate..I have got a 100 volt swing DC on across that vu resistor

 

[schmidlin]

On my (federal) units, Threshold and Release chase the meter around a bit, even at idle.  So, from my experience, this is normal.  Not sure about your "varying voltage and input" tho....

 

[college101]

Sorry schmidlin, I missed a word there...by adjusting the Input Volume pot- I can adjust where the meter falls...

 

[college101]

what do you guys think, put 1k test tone threw it...adjust the meter resistor till zero?

I re-checked my meter- wires connected in the correct polarity...

 

[MagnetoSound]

Right now playing with the controls on the faceplate..I have got a 100 volt swing DC on across that vu resistor

Yikes! ... there should be like a couple of volts or so at most.

I suspect the 6BC8 is not biasing properly.. The cathode circuit requires a DC current meter, so if you can remove the rectifier from inside the meter it will help. A standard VU meter will not work.

The meter should adjust to zero with NO signal. Sometimes the value of R8 must be fiddled to get this right, but since you say that adjusting the INPUT pot affects the meter, this strongly suggests that DC bias is present in the input stage during no-signal conditions, and you should really troubleshoot this before proceeding further. There is likely some grid leak current drifting around due to the high cathode voltage.

 

[college101]

Ok MagnetoSound, This is starting to make some sense...

My next move then is to remove the rectifier from the inside of the Honeywell...  This is a stadard Vu Meter I bought of Ebay..


Is it possible that the resistance of R8 is just too low? Maybe I need to move it up from the 34 ohms suggested on the schematic? I was geussing that it needed to be closer to the 100 ohms, but was waiting till I knew for sure before changing something from the schematic!

 

[college101]

SO, just to take some more measurements.. I checked the resistance across the honeywell vu meter...it came out to be 14K

I happen to have some other Weston vu meters laying around, so I pop them over into the circuit.
**Now I have two weston vu meters to try, The fist one is a 302- and it had 10.4K resistance...
  The second one, is a huge weston vu meter, that I dont have the model for, it measured 10.7...

  Once I put in either meter, the input volume control did nothing to meter movement. Now the release and threshold control still moved the meter a bit( probably a half dB of movement- but im not that concerned with that) The threshold is normally set a constant rate, and the release is adjusted to taste... The input is what I will be mainly playing with on the 436

What do you guys think?

Magnetosound.. I tried to take apart the Honeywell VU meter, BUT I will damage it if I go any farther.. I have take off the faceplate, but I cant see any internally connected diodes

 

[MagnetoSound]

The useful info is here:

http://www.groupdiy.com/the-lab/altec-436c-vu-meter/

Good luck.

 

[college101]

well..geuss im taking the whole meter apart now...Thanks MagnetoSound

 

[college101]

I have taken apart my vu meter...Found the fullwave recitifier circuit... How can I convert it to being a current meter?

 

[DaveP]

College101,

You're training to be an electrical engineer right?

Then let's have some voltages on your amp to reassure us something basic has not been missed.  I've attached the schematic of my amp to give you an idea of what they should be.

In your line of work attention to detail will be essential, so don't just stick any old meter in there and hope for the best!
You need to check out its parameters and carefully work out how much current goes through the meter and how much through the cathode resistor (avoids expensive mistakes).

Get this right and your tutor will be even more impressed if you managed to troubleshoot it yourself, no-one gets it right first time least of all me!

PS. 6BC8's are rarely balanced, they don't need to be in Cascode mode Magnetosound is right.  Thats why my circuit has a balance pot.
best
DaveP

 

[gemini86]

From the other thread... since it's relevent here..

Hello to everyone
I am realizing a clone of the Altec 436C Compressor and I'd like to know the specs of the original Vu-meter.
Any help?

Take a standard "real" VU meter and remove the diode bridge.  This will give you a 200 uA (microAmp) meter which is about what you need.  That's also what the Fairchild meter was. 

In the 436, the meter is in // with a 34 ohm resistor.  Make the 34R a rheostat (say a 50R pot) and you'll be able to trim it for the zero.

I've been inside lots of 436's and fixed quite a few of the meters.  I've used a new 100uA movement inside the case with no problem and adjusted the // resistor/rheostat/pot accordingly. 

The Altec meter doesn't follow the real amount of dB compression absolutely accurately (neither did/does the Fairchild) but, so what!  You can make it follow more precisely but it requires a different topology and is beyond your question.


Ciao.

 

[zebra50]

Also, I found this advice from PRR from a VERY old thread when looking for my measured voltages on my hard drive...

Re-biasing the 6BC8 will spoil the GR Meter calibration. They picked that 34Ω resistor across the meter so the static plate current in the 6BC8 brings the meter to the zero mark; with new bias it won't get all the way up to zero. Increase the 34Ω in very small steps (47, 68, etc) until the meter idles on the zero mark. Do NOT power-up without this resistor! The meter will blow. Be sure your test connections are solid. Don't use a trim-pot; wiper contact failure would be fatal to the meter. Don't be too fussy about the zeroing until the amp is fully warm. It does not have to read zero exactly, just twitch usably.

 

[college101]

Here is some data... on the 436

All readings were taken with input-threshold-release at min setting...

All info is given in Volts or (milli-Volts**in a few place)  DC

PS-
C8+=118.9
C7+/R13 junction= 237.8
C8b=215.03
C8c=146.01

From the Input --This was interesting?!
Green=-7.5
P1A wiper=-7.5
Middle connection=-7.5
P1B=-7.5
Yellow=-7.5

V1-6BC8
1=34.36
2=-0.848
3/8=127.3 mV
4=
5=
6=34.36
7=-0.85

High pass circuit
R3/C2=33.66
R4/C3=33.67

C2/R5=19.45
C3/R6=17.89

V2-6CG7
1=141.4
2=17.68
3/8=3.125
6=141.42
7=17.4 mV

R7+=3/8 of V2= 3.125

Red'white connection -Top of OT=140
Red-Bottom of OT=140

I said before I had a 100 volt swing...I completely missed the small mV symbol on my meter! 
The R8 adjustable pot has 56.6 mV with the input at min,  while at max it is the same
  ** must have had a bad connection when I took my measurement before... To explain with some more detail, there is no swing....

c4/R12 junction =-0.353
R12/R9=-0.353
R9/P2=.141 mV

V3  6AL5
1/R11=19.44
7/2=-0.353
5/R10=19.44

Let me know what you think guys

 

[MagnetoSound]

I think you need to get your volts and millivolts fully clarified before anyone else tries to decipher your readings ...

 

[DaveP]

Dan's right,
according to your figures you have between 17~19V of DC on your 6CG7 grids!
So is it -7.5V or -0.8V on your V1 grids?
Other voltages are ok.

This is how I would calculate for a meter using your figures.

B+ 215V Plate 34V, voltage drop=181V

181V/47K=3.85mA

3.85x2tubes=7.7mA Ik

0.0077x34ohms=0.26V Vk

So you need a meter which will give you fsd (inc the meter zero pot) with 0.26V

If as Gemini says you need a 200uA meter:-

0.26V/0.0002=1309 ohms meter resistance.

It's all just Ohms law stuff
best
DaveP

 

[college101]

Sorry guys for the confusion. I have to start over and re-take my measurements...

Can you point me to a good VU meter then? Should I go sifam?

The problem here is that, when I took my initial reading, the VU was connected...I was getting a voltage swing...most likely mVolts...and not volts...

I figured out my own mistake, and notated it wrong... I will be re-taking my readings again, and will report back...


I took out the Bridge rectifier in an attempt to see if the VU meter was able to be salvaged. I dont think its the right meter now based on the last few post from you guys...

Now when I measure across the R8.. There is no voltage swing now othern then for a split second, but thats because the VU meter is no longer connected...

Whats next...buy a new Dc Current meter...

 

[DaveP]

College,
I have attached the Altec meter scale.
As you can see, it's not a VU scale, its just a mA or uA meter with a decibel scale.
It is possible to modify a scale yourself, delicate job, but possible.
Sifam is top of the range but not really necessary, it's just a relative visual indication after all.
best
DaveP