Hello to all, and thanks in advance for your help and suggestions.
I’m getting ready to take a plunge into my first valve project (have done several SS based projects and repaired a few valve pwramps and preamps along the way). It appeared in the September Audioxpress. It’s an integrated amplifier that uses 6550’s in UL or triode mode depending on configuration. The design uses raw AC for the heaters. The author indicates that the circuit grounding scheme should not use a common grounding buss, but rather make all of the ground connections to the chassis, and as short as possible. Also the circuit and AC chassis ground are the same i.e. the aluminum chassis.
Question 1:
Is there a benefit to adding a bridge/filtering and using DC as the heater source instead?
Question 2:
Is this a common grounding method with valve circuits?
Questions 3:
Is there a benefit to placing the AC ground and the chassis 10 or so ohms away from each
other?
Questions 4:
What type of switch would you recommend I use to enable switching from UL to triode mode? I know most switches are current rated at 120/240 volts, and the B+ is upwards of 300 volts.
I’m getting ready to take a plunge into my first valve project (have done several SS based projects and repaired a few valve pwramps and preamps along the way). It appeared in the September Audioxpress. It’s an integrated amplifier that uses 6550’s in UL or triode mode depending on configuration. The design uses raw AC for the heaters. The author indicates that the circuit grounding scheme should not use a common grounding buss, but rather make all of the ground connections to the chassis, and as short as possible. Also the circuit and AC chassis ground are the same i.e. the aluminum chassis.
Question 1:
Is there a benefit to adding a bridge/filtering and using DC as the heater source instead?
Question 2:
Is this a common grounding method with valve circuits?
Questions 3:
Is there a benefit to placing the AC ground and the chassis 10 or so ohms away from each
other?
Questions 4:
What type of switch would you recommend I use to enable switching from UL to triode mode? I know most switches are current rated at 120/240 volts, and the B+ is upwards of 300 volts.
In a power amp there may be no reduction in hum by using DC heaters - AC should be fine. Preamps are more critical.
Multi-point grounding is fine as long as there is no current flow in the chassis - provide a return for each current path. NEVER NEVER NEVER run heater current or B+ return current through the chassis and 90% of the grounding issues are handled. There are many possible ways to achieve adequate grounding - which again, is easier in a power amp than a preamp.
In my opinion, anyone building AC powered equipment should be aware of safety requirements for commercial equipment and follow them. Any product must have exposed metal grounded, or be double-insulated. Since the latter is impractical, you must have a safety ground in the power cord. There is no current flow in the AC safety ground under normal conditions. It IS possible to break the ground with a pair of diodes, but iit should conform to the same test as a standard ground wire (less than 2.5V drop with 25A current).
A 250V switch should be OK if it isn't switched while power is on... then only the line-to-case insulation matters. The switch should be located where there is no possibility of "hot" switching - maybe under the chassis?
Multi-point grounding is fine as long as there is no current flow in the chassis - provide a return for each current path. NEVER NEVER NEVER run heater current or B+ return current through the chassis and 90% of the grounding issues are handled. There are many possible ways to achieve adequate grounding - which again, is easier in a power amp than a preamp.
In my opinion, anyone building AC powered equipment should be aware of safety requirements for commercial equipment and follow them. Any product must have exposed metal grounded, or be double-insulated. Since the latter is impractical, you must have a safety ground in the power cord. There is no current flow in the AC safety ground under normal conditions. It IS possible to break the ground with a pair of diodes, but iit should conform to the same test as a standard ground wire (less than 2.5V drop with 25A current).
A 250V switch should be OK if it isn't switched while power is on... then only the line-to-case insulation matters. The switch should be located where there is no possibility of "hot" switching - maybe under the chassis?
I thought this particular article was loaded with bad advice, (SORRY!) and 30yrs of design & construction experience tells me the author is all wet. You can get away with this in an AL chassis, but steel will have eddy currents induced from the power transformer that will modulate all of the grounds. Sometimes you can get away with this, but in general it degrades the amplifier noise floor. It's just BAD practice, and there is a good reason why people use buss and star grounding. Long term ground integrity is far better with one single mechanical grounding point with everything referenced to one point.
The measured performance of this design is not stellar either. The design is at best mediocre and there are much better designs out there. Both Eli and SY have published much better designs than this IMO.
Please go out and get yourself Morgan Jones "Valve Amplifiers" 3rd edition, and "Building Valve Amplifiers." He will make the reasons for my objections to this design very clear, and in a clearer, more concise manner than I ever could. There are also some very good designs presented in these books along with excellent explanations of many facets of tube audio design.
The measured performance of this design is not stellar either. The design is at best mediocre and there are much better designs out there. Both Eli and SY have published much better designs than this IMO.
Please go out and get yourself Morgan Jones "Valve Amplifiers" 3rd edition, and "Building Valve Amplifiers." He will make the reasons for my objections to this design very clear, and in a clearer, more concise manner than I ever could. There are also some very good designs presented in these books along with excellent explanations of many facets of tube audio design.
Relax! You will be fine. PLENTY of help is available here to put the stuff already in your possession to good use.
Is the article you refer to available online? I'd like a look see, but don't subscribe to any magazines. Please post a list of what you have, paying close attention to the magnetics and the tubes.
Well if you don't want to go with this design from the advice here, then there are always other options. What all have you purchased, I am not familiar with this particular schematic. Push/Pull or SE? I am assuming push/pull and will proceed thinking that.
You really shouldn't need DC for indirectly heated tubes like the 6550. DC is usually used in DHT tubes like 2A3's since the cathode is also the heater, therefore you need a very clean heater supply. I have kicked around a lot of designs to play with, I mean a lot. I have had a few threads here about SE amps that have kind of piddled out, mostly a lack of funding and time thanks to college. Recently I have really started to look at push/pull amps. Personally with the 6550's this would be my route:
6GK5, D3a, C3g, 6C45, 12HG7, something of this sort at the input.
12AT7, 5687, maybe 6SN7 as the driver/splitter (differential)
6550 outputs wired as triodes
Basically a somewhat modified Mullard circuit. Depending on what other tubes you have I am sure something could be rigged to work. Should be good for something in the neighborhood of 25watts give or take, maybe 30watts-ish. How much power are you needing and what kind of input sensitivity? There are tons of other options out there for push/pull that use or could use the 6550. There are also tons of people here that know way more than I, just my 2C worth.
Cheers
James
You really shouldn't need DC for indirectly heated tubes like the 6550. DC is usually used in DHT tubes like 2A3's since the cathode is also the heater, therefore you need a very clean heater supply. I have kicked around a lot of designs to play with, I mean a lot. I have had a few threads here about SE amps that have kind of piddled out, mostly a lack of funding and time thanks to college. Recently I have really started to look at push/pull amps. Personally with the 6550's this would be my route:
6GK5, D3a, C3g, 6C45, 12HG7, something of this sort at the input.
12AT7, 5687, maybe 6SN7 as the driver/splitter (differential)
6550 outputs wired as triodes
Basically a somewhat modified Mullard circuit. Depending on what other tubes you have I am sure something could be rigged to work. Should be good for something in the neighborhood of 25watts give or take, maybe 30watts-ish. How much power are you needing and what kind of input sensitivity? There are tons of other options out there for push/pull that use or could use the 6550. There are also tons of people here that know way more than I, just my 2C worth.
Cheers
James
Fortunately all of the iron is applicable to another design as is the chassis and a great many of the other parts.. Do go get the Morgan Jones books if you can. (I have picked up copies at Barnes and Noble)
I thought the driver stage in that design had surprisingly high distortion for the signal levels measured. That was one of my other objections.
The design was recently published in AudioXpress..
I thought the driver stage in that design had surprisingly high distortion for the signal levels measured. That was one of my other objections.
The design was recently published in AudioXpress..
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.