Choke? Where to begin?

[cogitech]

I want to add a choke to my Shuguang I-25 (EL34 P-P) tube amp.

Been trying to do some research but feeling less confident the more I read.

Where do I start? What specifications of my amp do I need to know in order to choose a choke with appropriate specifications? What are the critical specifications of a choke that make it suitable for one amp and not another?

Any help at all getting me pointed in the right direction would be greatly appreciated!

 

[TheGimp]

You can start with Peak input voltage at the cap before the choke, and current drawn by the amp at the output of the filter section.

You need to know the peak input voltage to insure the voltage rating of the choke is sufficient to insure it won't break down.

You will want to minimize the DC voltage drop across the inductor to keep from upsetting the design.

This assumes your intention is solely to reduce ripple voltage and hum.

 

[Rundmaus]

Simply adding a choke will not help.

A choke is usually a part of the filter network in the power supply, and should be understood as such. Not as an isolated 'add-on' that improves the sound.

For a first step, find a schematic or trace out the power supply by yourself. Then understand, with the help of this forum, how it works and whether a choke would improve it.

For playing around then, the PSUD II software will be helpful.

Andreas

 

[sreten]

Hi,

The choke for a CLC supply needs to be rated at maximum supply current.
For fairly obvious reasons they are more useful in pure class A designs
with a constant supply current rather than pushpull amplifiers.

As long as it doesn't saturate due to current any decent value
will help in a CLC supply, how much depends on how big it is.

rgds, sreten.

I'm assuming voltage rating is completely irrelevant, because it is.

 

[TheGimp]

Wrong on the voltage rating.

Inductors are specified for Maximum DC voltage rating.

They are not guaranteed not to break down above the rating. While most will be tested at higher ratings, it is a bad idea to ignore the specifications.

Check the Hammond site for an example:

Hammond Mfg. - D.C. Filter Chokes - (153 - 159 Series)

 

[cogitech]

You can start with Peak input voltage at the cap before the choke, and current drawn by the amp at the output of the filter section.

You need to know the peak input voltage to insure the voltage rating of the choke is sufficient to insure it won't break down.

You will want to minimize the DC voltage drop across the inductor to keep from upsetting the design.

This assumes your intention is solely to reduce ripple voltage and hum.

Thank you. The goals are to reduce ripple(?) and hum.

However, it is obvious to me based on your response and others that I am in over my head.

I have been told several times on various forums that I should install a choke.

Perhaps that is just hot air?

Perhaps a better approach is to describe exactly what I am hearing that I do not want to hear.

Low level hum: this is not too bad, and doesn't increase with volume. Easily tolerated.

Variable buzz: sounds like noise, sort of crackly, is variable, and increases with volume. After hours of hot running, starts to cycle (oscillate) slowly about once every 0.8 seconds. (is this ripple? RF?)

 

[Rundmaus]

No, I don't think this is hot air.

A choke in a CLC or other filter configuration can significantly reduce hum. But it is not a 'plug-and-play' thingy like some fancy cable.

A choke needs to be dimensioned according to the specifications of the amp and the power supply, and probably the supply needs some redesign to accommodate the choke.

Again, a schematic would be a good starting point.

Andreas

 

[tomchr]

Assuming your current circuit has a rectifier followed by a capacitor, then you can get some ripple rejection by adding the inductor/choke (L) followed by a capacitor (C) to ground.

The inductor will need to be rated for the peak current drawn by the load (the amp in your case). The peak current of a single-ended amp is roughly twice that of the current at idle. Recall, in single-ended class A, the current in the output stage swings from the idle value to near zero and up to nearly twice the idle/quiescent value.
You'll probably find that the inductor current is specified as the RMS value rather than the peak value. If my math is correct, for a single-ended amp, the RMS current will be roughly 1.71*Iq, where Iq is the quiescent current.
For push-pull amps, you'll have to find the peak output current from the datasheet of your output tubes. The peak current is usually way higher than the quiescent current.

Once you know the current and voltage requirements for your inductor, just pick the largest inductance you can afford or fit on your chassis. Keep an eye on the DC resistance (or ESR) of the inductor as you'll lose some voltage across this resistance. Note that plate chokes are usually fairly resistive, hence, not suitable as supply inductors. The Electra-Print 70 H plate chokes I have, for example, have about 1300 ohm DCR.

A less expensive solution with significantly better ripple rejection performance would be an actual voltage regulator for the B+ voltage. But that's a different topic...

~Tom

 

[cogitech]

Here is the schematic. I have no idea how to read it. Like I said. In over my head. Most of what I have read above is like reading Greek.

Full disclosure:

1) I have disconnected U5 (6.3v 1.2a) tap from the 4 pre-amp tubes and installed a separate 4a 6.3v transformer to power those 4 tubes, which allows me enough current to run 12BH7 in place of 12AU7.

2) I have replaced the stock .22uF coupling caps with .22uF Sprague 716P.

3) I have replaced the stock .47uF grid return caps with paralleled .22uF (.44uF total) Russian K-40-Y PIO.

[IMGHTTPDEAD]http://www.cogitech.ca/photos/speakers/I25_Sch.jpeg[/IMGHTTPDEAD]

 

[counter culture]

Perhaps you could post a picture of the wiring in the amp?

Schematic doesn't make sense.

I don't know how well-built these amps are (evidently not so well-built that they don't hum), but the actual wiring, particularly in a tube amp, can easily contribute to unwanted hum.

It could be that rearranging the wiring could improve matters without resorting to a choke.

 

[cogitech]

Schematic is only showing one channel... It is basically 12AX7 line stage, 12AU7 phase inverter, 2x EL34 power.

In its stock form:

[IMGHTTPDEAD]http://www.cogitech.ca/photos/speakers/i25inside.jpg[/IMGHTTPDEAD]

As I said, it does not look exactly like this any more, as I have replaced all those cheap yellow caps, as per my earlier post, and the 4 pre-amp tubes are supplied by a separate (external) transformer.

 

[rfengineer2013]

Vcelkamaja has a hum thread going on his SE kit that has a very similar front-end circuit.

The primary source of hum in his case was the filament AC coupling to the cathode in the upper 12AX7 input tube.

The voltage on the cathode in the top-section of V1 is about half-way between the top plate voltage and ground. In the other amplifier, the cathode was at about 155VDC and the heater was referenced to ground. Hum from the heater was coupling into the top cathode and was being amplified.

The primary solution in that case was to bias the heater up (I think he biased it up to 30V or so) and the majority of the hum was gone.

You might check out that thread - the heater bias circuit is described here: http://www.diyaudio.com/forums/tubes-valves/226191-hum-tube-amp-8.html.

Your amplifier is a bit different, it has a rectifier on the 6.3VAC and a 10,000uF cap - but no series filtering element, just the cap. It's quite likely that you are still getting some ripple from that rectifier/cap circuit, it would be 100 or 120Hz depending on your mains frequency.

The 12AX7 in that configuration has lots of gain - a little ripple can cause a lot of hum.

The filament supply looks like it floats (i.e. - not connected to system ground), so it's still possible to bias the filaments up closer to the cathode in your amplifier.

And I agree with the other posters - just putting a choke in the amplifier is probably not going to solve your hum troubles.

Check out the other hum thread - you might get a better idea of what is going on and decide what your next step might be.

Good luck!

 

[cogitech]

Note, due to the size of the replacement coupling caps and PI grid return caps, I had to get pretty creative with the connections.

Perhaps I shall take a photo of exactly what I have done and post it...

Don't get me wrong, the amp doesn't sound bad or anything. It's just that last night during a quiet passage I heard the variable "sizzling" and I'd really rather have it gone.

 

[cogitech]

Vcelkamaja has a hum thread going on his SE kit that has a very similar front-end circuit.

The primary source of hum in his case was the filament AC coupling to the cathode in the upper 12AX7 input tube.

The voltage on the cathode in the top-section of V1 is about half-way between the top plate voltage and ground. In the other amplifier, the cathode was at about 155VDC and the heater was referenced to ground. Hum from the heater was coupling into the top cathode and was being amplified.

The primary solution in that case was to bias the heater up (I think he biased it up to 30V or so) and the majority of the hum was gone.

You might check out that thread - the heater bias circuit is described here: http://www.diyaudio.com/forums/tubes-valves/226191-hum-tube-amp-8.html.

Your amplifier is a bit different, it has a rectifier on the 6.3VAC and a 10,000uF cap - but no series filtering element, just the cap. It's quite likely that you are still getting some ripple from that rectifier/cap circuit, it would be 100 or 120Hz depending on your mains frequency.

The 12AX7 in that configuration has lots of gain - a little ripple can cause a lot of hum.

The filament supply looks like it floats (i.e. - not connected to system ground), so it's still possible to bias the filaments up closer to the cathode in your amplifier.

And I agree with the other posters - just putting a choke in the amplifier is probably not going to solve your hum troubles.

Check out the other hum thread - you might get a better idea of what is going on and decide what your next step might be.

Good luck!

Thank you (to all, really)!

I am really struggling to grasp a lot of what you are all saying.

My plan is to live with it for the time being, perhaps reflow some of my cap solder connections... Also, since you mention the 12AX7 having high gain in this circuit, perhaps I will stick my set of 12AT7s in there and see if I notice any change with the lower gain factor (?)

I need to learn several orders of magnitude more than I know now about all that has been explained above before I can ever hope to start tweaking the circuit.

By the way, the power transformer in this amp runs very hot. Even after adding a big heat sink, it is hovering around 150 F.

 

[rfengineer2013]

I was poking around and found another thread showing your Orange Drop upgrade. Dang, you're right - those ARE big.

Anyway, I wanted to ask is this a "hum" problem or a "HUM" problem. The issue on the Hum thread was bad - it was a bonafide HUM problem that was the result of a poor kit design (also Chinese, BTW).

It involved much more than just adding a choke - in fact, a choke was tried early-on without success.

It required relatively major circuit changes to bring off. But in that case, it was worth it.

So, the question here is how much do you want to dig into this thing? If it's a HUM problem, it's likely the folks here can help you through it - but since you purchased this as a manufactured amplifier, I would think either there is a serious design issue with the amp -or- you got a lemon.

Hard to tell, but in either case the warranty (whatever it was) is probably void now. And I've heard returns to China are impossibly expensive anyway...

Where is this headed... Well, just to reinforce that upgrading and modifying commercial amplifiers can be a rewarding and educational experience. It's what we all enjoy doing, the building and testing and hair-tearing-out problems that come up.

But it's not quick nor easy in most cases, especially chasing down noise problems. And while the hood is up and the engine is all over the garage floor, the amp won't be playing music either (mixing metaphors and all that).

Up to you, but just fair warning that this can be a very addictive pastime and it will require a lot of perseverence to solve.

OK, back to regular programming...

(I was off writing this when your comment posted above. Good luck with the amp!)

 

[vinylkid58]

My plan is to live with it for the time being, perhaps reflow some of my cap solder connections... Also, since you mention the 12AX7 having high gain in this circuit, perhaps I will stick my set of 12AT7s in there and see if I notice any change with the lower gain factor (?)

I use JAN GE 5751's in my Jolida. Circuit is very similar to your amp, basically the same SRPP front end.

By the way, the power transformer in this amp runs very hot. Even after adding a big heat sink, it is hovering around 150 F.

Underspec'd for the current demands, as I'm sure has been mentioned already. The reality is it's a cheap amp, so shortcuts are going to be implemented somewhere.

jeff

 

[cogitech]

I was poking around and found another thread showing your Orange Drop upgrade. Dang, you're right - those ARE big.

Anyway, I wanted to ask is this a "hum" problem or a "HUM" problem. The issue on the Hum thread was bad - it was a bonafide HUM problem that was the result of a poor kit design (also Chinese, BTW).

It involved much more than just adding a choke - in fact, a choke was tried early-on without success.

It required relatively major circuit changes to bring off. But in that case, it was worth it.

So, the question here is how much do you want to dig into this thing? If it's a HUM problem, it's likely the folks here can help you through it - but since you purchased this as a manufactured amplifier, I would think either there is a serious design issue with the amp -or- you got a lemon.

Hard to tell, but in either case the warranty (whatever it was) is probably void now. And I've heard returns to China are impossibly expensive anyway...

Where is this headed... Well, just to reinforce that upgrading and modifying commercial amplifiers can be a rewarding and educational experience. It's what we all enjoy doing, the building and testing and hair-tearing-out problems that come up.

But it's not quick nor easy in most cases, especially chasing down noise problems. And while the hood is up and the engine is all over the garage floor, the amp won't be playing music either (mixing metaphors and all that).

Up to you, but just fair warning that this can be a very addictive pastime and it will require a lot of perseverence to solve.

OK, back to regular programming...

(I was off writing this when your comment posted above. Good luck with the amp!)

Just a hum (not HUM) problem. So not really a problem.

It was the variable "sizzle" last night that bugged me.

Home from work now. Wife has had it on all day. I cannot hear any sizzle now.

Perhaps the new Russian PIOs just had to settle in...?

I will listen intently overthe next few days to see if it returns. Fingers crossed.

 

[cogitech]

Just had another close listen. Intermittent sizzles or buzzes, but very infrequent (many seconds between very short zzzts), and seemingly random. I am thinking maybe RF from wireless network chatter...

Oh, and if I pause the source and wrap the volume pot to max, everything but the low-level hum disappears, and the low-level hum never increases in volume.

Will continue to monitor, but at this point my intuition is telling me that my sizzle issue is RF, exacerbated by the "antenae" on my huge caps.

 

[jjman]

If the amp was single ended I would say a choke would help. I tried going chokeless on my SE build but had to add one to eliminate hum. But you have a push-pull amp that cancels the power supply (and heater) output stage hum like a hum-bucking guitar pickup does. And it looks to me that there is adequate filtering to avoid PS hum in the PI and gain stage. I suspect a different problem/solution.

 

[TonyTecson]

I want to add a choke to my Shuguang I-25 (EL34 P-P) tube amp.

Been trying to do some research but feeling less confident the more I read.

Where do I start? What specifications of my amp do I need to know in order to choose a choke with appropriate specifications? What are the critical specifications of a choke that make it suitable for one amp and not another?

Any help at all getting me pointed in the right direction would be greatly appreciated!

a very good place to start... audiofilterchokes

and an electronic version...http://aquinaudio.webspace.virginmedia.com/act_choke.html

and here...http://www.diyaudio.com/forums/tubes-valves/214681-gyrators.html