Fisher 30-A performance differences among the three versions?

The attached schematics and parts list have been posted in another thread some time ago, but I thought I'd post them again to reference for asking a specific question about the three versions of the Fisher 30-A.

That question is especially relevant when noting the differences in capacitor values chosen for the 10000 series as compared with the 20000 and 30000 series 30-A amplifiers.

I am wondering if one can tell or estimate what the impact is to frequency response, or any other performance measure, for the 10000 series as compared to the later two series.

You will note specifically the difference in value (.022 versus .047uF for example) in the C1, C2, C6 and C8 capacitors between the three different serial number series.

It's also interesting to note that the 10000 series seems to have at least one additional capacitor, C4, and the 30000 series has an additional or reconfigured resistor, R16/R10.

Is one design better than the other? Would they perform differently?

I have the 10000 series, with the single 2.5Ohm output transformer and the values in the spec match with what I have in mine.

Thanks,
Al

Little -- The different components are all due to the change in output transformers between the various units.

While you could likely get away with increasing your coupling caps (C1 & C6) to .047 uF in your unit, the changes with C2 you would not want to make as they are very OPT specific. C8 forms a DC block and high pass filter together with the input control, and is unchanged between the various versions. The increase in C1 & C6 is likely due to improved LF performance in later version OPTs.

The change with R10 & 16 simply allows for including an AC balance control into the design. If you have the means to adjust this, then it could be added to your version as well. Otherwise, I would leave that aspect of the circuit as is.

Dave

Thanks and very interesting observations.

So, when you say I might "get away" with changing C1 and C6 to .047uf, which *may* improve LF response, what might be the issue/risk with doing so?

I really hadn't planned to make any changes, but that sounds interesting.

Thanks again,
Al

With all else equal, increasing the size of the coupling caps increases the open loop gain (OLG) of the basic amplifying block at the lowest frequencies -- the loop being the negative feedback (NFB) loop ("open" means with the loop disconnected). With more gain at lower frequencies then, that means there will be more feedback at lower frequencies as will with the loop closed if these caps are increased in size.

Increasing feedback can have its advantages in lowering distortion, and improving speaker damping. But if it is taken too far, the amplifier can become unstable at low frequencies, producing anything from "breathing" woofers with no signal present, to muddied bass performance when a signal is.

To prevent such problems, designers tailor the low frequency response of the basic amplifying block to actually reduce feedback at the very lowest frequencies to maintain stability under feedback conditions. This is usually accomplished by carefully choosing the values of coupling caps in conjunction with following stage grid resistors to produce the desired roll off. Indiscriminately increasing the value of these caps without being aware of the concerns involved can produce and unstable amplifier, where it used to be completely stable.

If you do increase the size of your coupling caps, just be aware of the concerns involved so you make sure they are not adversely affecting the performance of your amp.

Good luck with it!

Dave

little-al,
The first amp has an output transformer with only a 2.5 ohm tap on it. There's one difference. Some of the later changes might have had to do with keeping the amp stable into a different output transformer.

IIRC the 30000 series 30-A was used in the 560 Series "COMPANION" in 1959 and 1960 and possibly was designed with a specific speaker setup in mind. The 10000 and 20000 were possibly sold as separates, in 1959(10000) and 1960(20000). The 30A in my 560 has a 39xxx serial.

Larry

dcgillespie said:

Little -- The different components are all due to the change in output transformers between the various units.

While you could likely get away with increasing your coupling caps (C1 & C6) to .047 uF in your unit, the changes with C2 you would not want to make as they are very OPT specific. C8 forms a DC block and high pass filter together with the input control, and is unchanged between the various versions. The increase in C1 & C6 is likely due to improved LF performance in later version OPTs.

The change with R10 & 16 simply allows for including an AC balance control into the design. If you have the means to adjust this, then it could be added to your version as well. Otherwise, I would leave that aspect of the circuit as is.

Dave

Click to expand...

Thanks, I'm really curious about the "R16" potentiometer on the 30000 series. You mention that is an "AC balance control". Can you expand upon what that accomplishes? It looks like in the schematic that they've simply added the 50k potentiometer between R10 and ground.

Could that 50k potentiometer be added to the 10000 series as simply as installing it that location?

Would there be any benefit or downside?

More than anything else, just curious what purpose it serves compared to the 10000 and 20000 series which do not have that pot.

Thanks,
Al

So, if one wanted to add the hum balance control that the 30000 series has to a 10000 series unit, what values such as resistors would need to change? It looks like R2 and R10 would need to move the 30000 series spec. Any others?

Little -- To understand the benefits of the AC balance control, you have to understand the basic operation of a push-pull amplifier in the first place.

Briefly, the idea of push-pull operation is to take the input signal and besides amplifying it, create two signals from it: one which is an exact copy of it, and one that is a mirror image of it, or more technically, 180 degrees out of phase. These two signals that are 180 degrees out of phase ultimately drive the push-pull output transformer -- one signal driving one end of the primary winding, the other signal driving the other end of the primary.

This connection provides many benefits that range from increasing power output, to causing many distortion products to cancel out within the transformer action.

The point of the AC balance control is to make the two opposing signals as equal in amplitude as possible, so as to create as much distortion cancellation in the OPT as possible. Since many variables of circuit operation can affect that balance, a control is provided in better designs to adjust for best overall push-pull performance.

Setting the control most accurately requires distortion measuring equipment, but other methods are also available that can help to target a very good setting as well.

I hope this helps!

Dave

On a related note Dave, I have always wondered about this. if I turned the inverter balance pot all the way to one side, would the 30A be functioning like a a single ended amp rather than push/pull?

HiFi -- The AC balance control does not have enough range to completely shut the signal off to one tube, but you generally wouldn't want to do that anyway.

In a single ended amp, the single output tube must be biased into Class A operation, so that the tube conducts current at all times regardless of the level of signal applied to the tube. In the 30-A however, each output tube is biased for Class AB operation. This mode of operation has each output tube conduct current at all times as well -- but only up to a very limited level of power output (usually only a couple of watts), after which, each tube is alternately "cut off" for a significant portion of each cycle of the signal applied to the tubes. When this happens, the stage has shifted into Class B operation -- hence the overall classification of Class AB. As a result, while the amplifier would in fact operate with signal supplied to only one tube, it would run into distortion rather quickly as the active tube moved into the Class B portion of its operation -- this because it would then only be amplifying a portion of the signal rather than the complete wave form.

Also, single ended output transformers employ what is known as an "air gap" in the magnetic circuit, which prevents the transformer from saturating from the heavy DC current that is always flowing in only one direction in such a transformer. Because of the way a push-pull output transformer is wound however with its center tap connection, the current flow in each half of the primary is effectively opposite that of the other, so no such air gap is required. Therefore, even if you left both output tubes installed to balance the current, driving only one tube would cause the current to quickly become unbalanced again, causing the transformer to saturate, and further adding to the biasing problems mentioned above.

Trying to make a single ended amplifier out of a push-pull design has surely been tried, and does produce a limited form of operation. However, the best single ended amplifiers are the ones that start out life that way in the first place!

I hope this helps!

Dave