[cooperpwc]

Like the title says. I know that rebiasing is important when switching tube types (i.e. between 12FK6, 12AE6 and 12FM6). My Millett Hybrid came biased by dizzyorange to 13.5 volts for a pair of Dumont 12FK6. Recently the sound seemed a bit worse than before so I tried some other 12FK6s: RCA and then Tung Sols which I am currently listening to. SQ is right back in the zone - excellent in fact. I haven't rebiased though.

I have a cheap multimeter and have taken apart the case before and found the bias points so I could. I know that Headroom has a three way switch for the three tube types on their special HeadRoom Desktop Millett but otherwise state "[the radio tubes] operate over a very wide voltage range. Therefore, there is no need to bias each tube individually or ultra-precisely". Is this true for the standard Millett Hybrid? Should I open it up and fine tune between 12FK6 tubes? Thanks in advance for feedback.

 

[tomb]

Yes. I'm not sure, but the Headroom version may include a type of auto-biasing circuit.

The DIY revMH Millett Hybrid has none of that. I have seen different tube types off by as much as 7 or 8 volts - that's over 50% of the bias range. Even in the same tube type, the typical tube is still off by 2 or 3 volts.

A cheap multimeter is all it takes, but be careful where you stick those probes. The ground test point is directly adjacent to the tube bias test point and it is easy to short the tube bias to ground if your probe slips. You can touch the ground probe to a ground point anywhere else on the amp - including the negative power connection or the signal input ground.

The revMH Millett Hybrid should run anywhere from 24 to 30VDC. Each tube's bias should be set to one-half of the supply voltage. A common supply voltage is 27VDC, so the tubes would be biased at 13.5V each, for 24VDC, they should be biased at 12V each, etc., etc.

 

[dsavitsk]

Quote:

Originally Posted by tomb /img/forum/go_quote.gif The revMH Millett Hybrid should run anywhere from 24 to 30VDC. Each tube's bias should be set to one-half of the supply voltage. A common supply voltage is 27VDC, so the tubes would be biased at 13.5V each, for 24VDC, they should be biased at 12V each, etc., etc.

This is a rule of thumb, but not critical. The difference between the bias and B+ simply defines how much voltage is dropped across the load and thus how much room the tube has to swing. So, biasing to 18V with a 20V PS will only give 2V of swing which means distortion. But, biasing to 13 or 14V with a 24V supply is fine. And, in fact, the tube has no idea what the B+ is, so biasing to 13.5V with a 24 or a 27 or a 30V supply should sound the same. Any differences are differences in how the CCS responds to the voltage drop, not in the tube.

That said, you want the tubes to equal each other for best performance. But, if they are super mismatched, then even though boas and current are the same, mu and Gm may be much different. You have no way to know this, though, so equal biasing is the best bet for matched channels.

 

[kramer5150]

I run my tubes at 12V, with a 24V P/S. IMHO the sound doesn't change at all... 10V-14.5V At least with my millet.

 

[cooperpwc]

Thanks for the feedback all. Okay, I'm going to open it up and rebias. Concensus on hear and DIY forums seems to be 12-14 volts as the standard range (although n_maher biased his to 17 volts with a Yound-Jung PS that runs at 30V). My multimeter was dirt cheap and not all that accurate in absolute terms anyway. The important thing seems to be to balance the tubes.

 

[tomb]

Yeah, believe me - the other posters are technically correct - but I've seen swings of more than 10V in bias between tube types (such as 12AE6 vs 12FK6).

 

[cooperpwc]

Quote:

Originally Posted by tomb /img/forum/go_quote.gif Yeah, believe me - the other posters are technically correct - but I've seen swings of more than 10V in bias between tube types (such as 12AE6 vs 12FK6).

Indeed. One individual had a swing from 13.5v with 12FM6 to 20v with 12AE6A. I am staying with the 12FK6 for now but balancing the tubes seems to be a good idea since I have switched pairs. I certainly can't see any downside.

 

[jamato8]

Bias of power tubes has a big effect on sound and some tubes just never come into their own unless the bias is right. The 90 needs to be run hot and it sound fantastic but run it like you would an EL34 or KT88 and you may be happy but you will never realize its potential. Dynamics are lost and the bloom just isn't here. Some tubes are more forgiving but there is no reason to try a number of different biasing points that are within specs.

 

[cooperpwc]

Update: It turns out that it was good advice to rebias my Millett Hybrid. I switched to the RCA 12FK6 tubes, same type as the Tung Sol and the Dumonts before. Whereas the amp was biased to 13.5 volts for the Dumonts, it was under 10 for each of the RCAs! I rebiased to 13 volts on each side and the RCAs, which seemed not quite as good as the Tung Sols before, now sound excellent.

This time I reassembled the amp with the top off. The top actually still fits over the tubes but now just resting atop the case instead of slipped into place and screwed securely. I have 14 pairs of tubes to play with (thanks dizzyorange). I avoided this for the first couple of months but I have a feeling that the experience of tube rolling is a bit like eating peanuts - hard to stop after just one.

 

[dizzyorange]

Cooper, glad to hear you're enjoying fiddling with the amp. I heard a fairly noticeable difference between bias voltages on this particular amp. One thing you can try is to listen to one channel only, and change the bias for that channel as you listen. The difference will be apparent then (subtle, but noticeable).

Basically, I found that higher voltages gave a narrower but more focused sound, while lower voltages made the soundstage seem wider but perhaps a bit fuzzy (of course you need to hear both channels to get the soundstage effect).

The lower voltages, say around 10.5 to 11V also seem to give a bit more of tube bloom, or "creamyness".

 

[dsavitsk]

Quote:

Originally Posted by dizzyorange /img/forum/go_quote.gif Basically, I found that higher voltages gave a narrower but more focused sound, while lower voltages made the soundstage seem wider but perhaps a bit fuzzy (of course you need to hear both channels to get the soundstage effect).

Generally with triodes, the higher they are biased, the lower the THD, but the higher the odd harmonic distortion. The lower they are biased, the lower the 3rd, but the higher the second harmonic distortion. Thus, the lower the bias, the more bloomy and tubey it will sound. The higher, the clearer but also sharper and harder it will sound. Finding the right operating point is an exercise in balancing these positions. Depending on the tube, these differences can by small or large (and I have no idea for the Millett tubes other than to say they are not particularly linear in the first place, but it is probably mostly even order distortion.)

 

[cooperpwc]

Thanks dizzyorange and dsavitsk. I expect to start seriously playing around this coming weekend. That gives me some guidance.