I'm thinking about having another go at my F5. I've seen various schematics floating around. The “official” one in the F5T manual shows to use 100Ohm gatestopper before Q3 and Q4. However others show gatestoppers of 49Ohm. The use of gatestoppers are said to degrade the sound quality, so presumably a lower value is therefor preferred. But I suspect the 100Ohm value is chosen to be on the safe side of most eventualities.
My questions are, is it worth changing these out to 49Ohm (compared to 100Ohm) and what’s the risk of oscillation becoming a problem? Will the gain be higher with 49Ohm’s?
My questions are, is it worth changing these out to 49Ohm (compared to 100Ohm) and what’s the risk of oscillation becoming a problem? Will the gain be higher with 49Ohm’s?
Not anywhere as much as you might think.
Possibly.
No change in gain.
Gatestoppers on a mosfet amp can be anywhere from 47 ohm to about 680, where a higher value will do more to damp ultrasonic ringing (I.E., oscillation). Use whatever you happen to have. I like about 150ohm personally.
The F5 has low gain and no stability compensation.
The gate resistor may well be a mild form of stability compensation.
Since no other compensation is used, I would consider the PASS value to be optimum and am wary of changing that.
100ohm is easier to read than 100Ohm.
But I prefer 100r for resistors. Very easy to read and understand.
ohm can be used for resistive, or reactive impedance and can be confusing in some contexts, so when referring to resistive, I always use the r, k, M, m as in 4r7, 2M2, 100r, etc.
The gate resistor may well be a mild form of stability compensation.
Since no other compensation is used, I would consider the PASS value to be optimum and am wary of changing that.
100ohm is easier to read than 100Ohm.
But I prefer 100r for resistors. Very easy to read and understand.
ohm can be used for resistive, or reactive impedance and can be confusing in some contexts, so when referring to resistive, I always use the r, k, M, m as in 4r7, 2M2, 100r, etc.
Last edited:
You can try this experiment with gate stoppers -- as demonstrated in one of the TI (was Nat Semi) application notes -- it think the author was SpittinLama -- use 100R on the N-channel, and a 500R pot on the P-Channel. Inject a square wave into the amplifier and observe the output. adjust the trimpot until the poz and neg waveforms are symmetrical.
Square wave will tax the amplifier so 70% is probably as high as you want to go.
Square wave will tax the amplifier so 70% is probably as high as you want to go.
Thanks for the extensive answers. I never suggested doing away with the gatestoppers alltogether. J
Based on your feedback I’ll leave the 100r’s in for now and concentrate on other parts of the amp. Especially reading Bob53's account of the amp going ustable with 47r's.. (sorry to hear this)
Side stepping from the original title of this thread, these are some other changes I’m thinking about, maybe some of you are willing to comment on their pro’s/con’s.
1. Changing R1&R2 from 4.7K/47K, back to the original 1K/100K. I’m not sure why this was ever changed. (higher input impedance seems better)
2. Ideally replace R5/P1&R21/TH1 with one HQ resistor (idem other side), after burn-in period when bias has settled in. (improve supply feed to input stage)
3. Change the feedback voltage dividers (R9-R12) to 100r. (Gut feeling tells me giving the JFET's a little less feedback from the heavy hitting Mosfets, might be a good thing)
Based on your feedback I’ll leave the 100r’s in for now and concentrate on other parts of the amp. Especially reading Bob53's account of the amp going ustable with 47r's.. (sorry to hear this)
Side stepping from the original title of this thread, these are some other changes I’m thinking about, maybe some of you are willing to comment on their pro’s/con’s.
1. Changing R1&R2 from 4.7K/47K, back to the original 1K/100K. I’m not sure why this was ever changed. (higher input impedance seems better)
2. Ideally replace R5/P1&R21/TH1 with one HQ resistor (idem other side), after burn-in period when bias has settled in. (improve supply feed to input stage)
3. Change the feedback voltage dividers (R9-R12) to 100r. (Gut feeling tells me giving the JFET's a little less feedback from the heavy hitting Mosfets, might be a good thing)
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.