• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

ST70 hybrid finally up and running !!

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Finally got my ST70 (or what's left of it) running. Basically, I took the xfmrs out of a ST70, and built everything else from the ground up, chassis, power supplies, you name it. Had a ST70 many years ago, so I wanted to relive the glory days, but kind of take it to the extreme.

Haven't had too much auditioning time, but I have some 60Hz noise going to the speakers. It's not excessive, and doesn't bother me at all, but I am a little miffed as to its source.

It's not 120Hz, checked with a scope. Level is about 50mV peak-to-peak with a 4 ohm speaker connected to the 8 ohm tap (long story-don't ask). Output B+ has 80 mV ripple, driver stages less than 5 mV ripple.

All heaters are DC, regulated with LM317 as current source. Star grounding throughout, and only one connection to chassis (IEC ground) through a 100 ohm resistor. If that resistor is lifted, I get open circuit with Fluke 87.

Ideas on where this hum is coming from? I can submit schematics, but will need a little time to post. Basically 6922 cascode input followed by 6922 concertina, driving EL34's.
 
Here's what were dealing with.....
 

Attachments

  • st70.jpg
    st70.jpg
    70.1 KB · Views: 638
Hum?

Hi,
Great work!!! I never have the patient to build such handsome looking chassis.

2+1 places I normally look into when hum come out from fresh amp,
1) Layout of the wires, make sure all wire crossing are perpendicular to each other, avoid wires running in parrallel, and twist all AC wires.
2) check if any small signal wires are close to PSU or other RFI sources in the chassis...
+1) if 1 & 2 fail, increase thepower supply's first capacitor size.... :smash:

As for the grounding scheme, my personal preference is to separate the ground of the power supply from the signal circuit... at least the 1st cap's ground is connected directly to the ractifier's/diode's/bridge's ground pin. and I like to separate the ground for small signals from bigger signals...
 
Paid very close attention to wire layout. There is a star point for input stage, then a star point for output stage, and a star point for power supply. These get tied together in one place.

What is most confusing is the relative absence of 60Hz carrying wires inside this amp. They are confined to two areas, and are run with at least 3 inches clearance from all significant circuit paths.

I used a distributed PCB layout of power supplies, followed by point to point wiring of the tube sockets. All heaters and power supplies, even though DC, are tightly twisted pairs.

I will try lifting vs shorting my 100 ohm resistor to chassis, see if that makes any difference.

I've got quite a bit of capacitance already, and like I said, my noise is not 120Hz. It's either got to be coming in through the ground point, or possibly coupled from the power xfmr.

Source for testing is an iPod, so no loops there.
 
zigzagflux said:
possibly coupled from the power xfmr.

If it's the power xfmr influencing the OPT's, then you will hear the hum loudest right at turn on (current draw, hence flux, is greatest then) - even before the tube filaments are warm and the amp is amping. In fact, you can try it with no output tubes.

Another easy check is to take a plate of metal (steel is good, but thick aluminum is also fine) and slide it between the power xfmr and OPT, then listen for a change in hum.

Also, if you have a steel chassis, try isolating the transformer from the chassis by a mm or so, with non-conductive spacers.

Sheldon
 
Good suggestions, but already done, my friend.

Power transformer is electrically and mechanically isolated from the top plate through rubber mounts and teflon shoulder washers. Single point ground to chassis from there. Didn't want mechanical vibrations to be an issue.

I use a delay circuit, such that tubes are heated before B+ is applied. Hum only starts after B+ is up, so I guess the chances of xfmr coupling are reduced? I do know that the physical spacing on my top plate is greater than that found in the original ST70 chassis.

I will pull the tubes tonite and see what happens. Will also poke around with the o-scope to try and find the leakage source.

Thx
 
Well at least I've had some progress.

As far as I can tell, the 60Hz is entering at the passive input stage. I will be using a powered subwoofer, so have a first order BP filter at 90Hz and 50 kHz prior to the first amplification stage. The LP at 50 kHz was based on Van Alstine recommendations.

When the RCA input is shorted, there was essentially very little reduction in output hum. However, when the input grid was shorted (post-filter), All is Quiet on the Western Front. So the filter is picking up 60Hz EMI.

There was a 120V twisted pair run within say 2 inches of this passive filter, which surprisingly was a partial culprit. I was surprised because this pair feeds only my bias circuit, which draws essentially no current. Current carrying conductors produce stronger fields than only voltage carrying, but it is what it is. Moved it to the other side of the case, and cut my hum by about 50%.

The rest of the hum can still be seen on the scope, but is very difficult to remove. There is literally nothing nearby. The closest transformer is maybe 4-5 inches (sorry, 10-13cm) away, and is small signal.

My next choice is to remove the LP filter entirely, and decrease the resistance of the HP filter (and increase cap accordingly). This will hopefully reduce hum even further.

Ultimately, the level is very low, about 3mV rms at the speaker terminals, but it bothers me that its there, and it can be heard with the ear up to the speaker.

Anyone have good results with the Van Alstine recommendations?

http://www.avahifi.com/root/equipment/amplifier/70_rebuild_plans.htm
 
I have learned a number of things during this project, and will assemble most of my future amps a little differently.

B+ and filament supplies will be mounted in a separate chassis. Not only do I get the sonic benefits, but I can physically fit more stuff into each respective chassis. That way, I can add additional filtering / shielding / spacing where needed.

It is amazing how adding a few extra features constrains you. I have ten pounds of schmutz in a 5 pound box.

Not only that, but I could use the power supply for potentially a number of different amplifier projects.

I have pretty much resolved all my hum in this amp. The problems were strictly localized EMI, addressed by routing AC conductors very far from the input stage. Next time, I will consider shielding, as suggested. Surprised how sensitive an input stage is !

All grounding, as far as I can tell, was acceptable. My next problem was RF amplification, which was addressed in a separate thread. Going with a different driver stage to hopefully remedy this. Parts come in Friday, I may have an update by Sunday night.

Thanks everyone for your suggestions.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.