stay at home type 69 project

[fritzmyname]

All my pcbs are stuffed i just only wired one of the front panels cause I wasn’t sure where to put the stop pins. I’m not so sure the second stop in is the same for the EQ and the gain. Gain is 6 positions only and for that I used 12 o o’clock and 6 o’clock and it stops correctly. For the 11 position eq switches I believe it could be what twenty trees posted. Can you confirm ?

I left the pins for now so I cannot really tell you

Could you maybe post some pictures of the unit so we can compare?

 

[dogears]

It's definitely a bonding experience that we're not alone in this haha

@dogears did you change anything in the PCB file? Could there be anything that the PCB plant screwed up because we are too inexperienced when ordering PCBs?

Could there be any parts that we screwed up ordering? Can you check the BOM? Maybe @dogears can identify some critical parts that we maybe screwed up ordering? I for one used the Cinemag transformer and had to resort to slightly different C values when ordering.

As far as I remember @TwentyTrees had a nother solution for the 24V regulator circuit than we have in the current PCBs. Could there be a screwup?

I just checked the pictures on the IAA website and the design is not exactly the same for the H9. Could you identify the problem areas that would cause such issues? It seems we all deal with the same problem here so maybe we can find where there is a problem. Would it be an option if you checked the current PCB version yourself (meaning produciing one to verify that the GERBER file is not the issue)?

good questions.

well, twentytrees made a couple of versions of this unit. First was one that had +/-16V total, which both of us discovered had an overheating problem. Always thought that was a bit odd as there are multiple schematics floating around out there with these units running on higher voltages. after we discussed this he installed the regulator to drop the +V rail to +8VDC putting us back to +24V effective for the circuit. I revised the H9M files for this approach. He built a few units in a configuration I made for him, which is completely identical to the current save only a different transformer footprint and he wanted to try tantalum caps (this is H9MA).

the only difference between the original H9M and the revision is the installation of U1. And we've confirmed that the voltages are correct, so I assume the 7808 is working.

the transformer change shouldn't matter - there seems to be an issue around the second amplifier block. but that is 100% identical to the first amplifier block, and works off of the same voltage rails. the only difference is the second has a fixed gain set resistor instead of selectable via the gain switch.

this is my confusion - if the first gain block works, and the second doesn't... ????? it's the same parts, running off of the same rails. there has to be a reason or a difference, but i can't find it.

there are a couple differences between the production version, but the amplifier parts and values are identical to what is here. so - maybe someone can find the difference between amp 1 and amp 2.

 

[Janalex]

This morning so far I soldered my extender to externalize the module from the rack. Also I had the gain stop pins in wrong. Used a magnet to remove them and now I can confirm this is the correct stop pin arrange for the gain switch based on checking the resistors with a multimeter. I will use the twenty trees pic for the other switches as that makes sense intuitively have the pins right next to each other. I will test shortly and report back.

 

[dogears]

for the grayhills - set the first pin at 12 oclock, then the second pin at the number of stops. so for six position, 12 and 6 oclock is the correct setup.

 

[Janalex]

So after correcting the pins and testing I have a proper working gain switch. I am now hearing my voice with a mic but there is increasing buzz saw distortion as you turn up the gain. Even with the distortion I can tell the mid and bass gain are functioning. Pad is off. First thing I’m noticing is 48v is always present on pins 1 and 2 regardless of switch position. I will post pics. The only mods were removing the caps C51 and C2 replacing C51 with the 47pF suflex cap and shorting R54. Also RL1 removed. I can test some voltages today and report back if anyone has any input. Tomorrow I will wire the second of my 4 boards to test without those front end mods and see if same thing occurring.

 

[Janalex]

Actually the phantom voltage was likely persisting due to the pc1 cap. If I wait it will likely dissipate.

 

[Janalex]

Preliminary voltage readings with the extender card yield vastly different numbers than the ones posted here. I also connected another channel that didn’t yet have wires to the front panel or transformer and while the numbers weren’t exactly the same as the completed channel the abnormal patterns were the same like pins 1 and 3 on the transistor being the same on q7 for example. I will post the exact numbers this evening as I have a session soon. Long story short voltages are low.

 

[dogears]

thanks Janalax. The EQ functioning makes sense since previously people were saying the first amplifier was working correctly, and the EQ is passive (lossy). You could verify by wiring in an output directly at the EQ tap point / entrance to the second amplifier.

I am back to wondering how the two voltages could be different on the two amplifier blocks as they are identical. Makes me wonder if there is not a hidden problem in the pcb file, like a short not caught by DRC error.

 

[Janalex]

Here are my numbers. My 2 sides of the amp are not significantly different at all which is a good thing. However the numbers are off.

Q1 -2.8 -3.4 -2.8
q2 +12 -2.7 -3.3
Q3 -4.1 -3.3 -4.1

Q5 -2.8 -3.3 -2.8
Q6 +12 -2.8 -3.4
Q7 -4.2 -3.4 -4.2

Emitters and collectors the same value in 4 of these?

And like I said I saw the same pattern repeating in another channel without wires to faceplate so I made a mistake and repeated it 4 times. Any ideas ?

 

[TwentyTrees]

I've just managed to find my notes on the first H9M I built, which as noted earlier did not have the regulator onboard (so I used a KA7908 on the -16V rail). My transistor DC voltages were apparently (pins 1-3 by schematic):

Amp 1:

• Q1: 1.33V; 0.77V; -2.37V
• Q2: 13.9V; -2.37V; -2.95V
• Q3: 2.59V; -2.95V; -3.6V

Amp 2:

• Q4: 1.5V; 0.93V; -2.1v
• Q5: 14.2V; -2.1V; -2.7V
• Q6: 2.8V; -2.7V; -3.55V

I agree @Janalex, your voltages look unexpected in comparison - specifically the pin 1 side. Are your positive and negative rails (pre and post any regulator) where you'd expect?

 

[dogears]

well - ok - the pcb rev I did uses a 7808 to drop the +V rail to +8 rather than a 7908 to bring the -V rail to -8.

not sure why this would be an issue either way. but the voltages would need to be shifted accordingly.

 

[TwentyTrees]

Agreed, can't see why that would make a difference. I only went with 7908 on V- as there was a bit more space to play with on that part of the PCB. So presumably, add 8V to my voltages above for full rail-to-rail measurements.

 

[Janalex]

I will check the rails tomorrow on this one and wire up another and see what happens.

 

[Janalex]

Hey guys,

So I wired up a second module and still not getting what doggears posted. I didn’t do any mods to this one nor even wire in the transformer. Can we just break down one spot in the circuit to start? Maybe it will offer us a clue.

Coming off the -16v rail to pin 3 of q1 for example. R9 220k brings the voltage down to -4.0v for me. You have -9.3v. How can this be? Resistor is correct value. Then I have the same -4.0v on pin 2 of q2. R14 brings the -16v to -4.6v and it’s the correct 22k. So that’s what I have at pin 3 of q2 and pin 2 of q3.

Off the positive rail right after r23 I have +5.0v which is what q2 pin 1 sees. Again you have almost 7v.

Why are my voltage drops higher? Could it have anything to do with the orientation of the transistors or something else ? Fritzmyname can you mesures these spots?

 

[dogears]

LTspice sim, should be reasonably accurate.

 

[Janalex]

Thank you. However I’m no electronics expert but what may be the cause for example of one side of R9 for example being -16 and the other side be -4V on my board if the resistors in that area are all the correct value? What am I missing or should I check? And this is happening in several spots.

 

[dogears]

I don't know - but if you figure the working +v rail after R23 being +6 and the -V rail after the diode being -15 or so, -4V is around the middle between them. Seems relevant not sure why. Transistor pinout would be my fist thought, but we've been down that path.

 

[Janalex]

I’ll check the pinout again. I have 2 more modules I can also run tests on. Question: would you expect the voltages on the transistors to be the same before and after wiring the pots and switches on the front panel? My last 2 are not wired but I want to measure those as well to compare.

 

[dogears]

yeah, it shouldn't change anything. the amplifier blocks are all biased without anything else going on.

one thing to try as a sanity check would be to remove the 7808 and short across it (or replace it with diode) and install a 7908 where D2 is.

 

[oongbac]

yeah, it shouldn't change anything. the amplifier blocks are all biased without anything else going on.

one thing to try as a sanity check would be to remove the 7808 and short across it (or replace it with diode) and install a 7908 where D2 is.

It works!!

I moved D2 to position of original regulator(7808) and put 7908 to original D2 position like picture above.

I just tested for a minute so there can be some long-term risk but it works for now anyway.

Thank you very much everybody who pointed this is not just a problem of mine and helped me went through this!

Please point out if there's some problems.