Eico HF-22 rebuild/update

ab0ez

AB0EZ
Well here I go again! This time it's a pair of Eico HF22 monoblocks that are in really nice cosmetic shape.

They had a half-hearted rebuild some time ago, but there are some problems with that. For one, the can caps aren't really soldered to the chassis at all, so any good low impedance ground connection is probably more of a stroke of luck than anything else!

Looks also like most of the original resistors were left in the amp, even though some show signs of thermal stress. They're gonna get round-filed and replaced with components hand matched to less than 1% tolerance.

All the old caps will be replaced with hand-matched items.

The power supply will get a choke added. This looks to be the 22A version that doesn't have a whole lot of filtering on the B+. We'll fix that! Along with a lot more filtering, I'll probably use a bus ground vs the chassis ground.

The "death cap" on these amps are just that. Not rated for this use and will be drop-kicked into the trash can.

With all these being said, does anybody have any hints or suggestions for other improvements, particularly the front end of the HF-22?

More to come as parts arrive.
 

Attachments

  • DSCF6834.jpg
    DSCF6834.jpg
    79.9 KB · Views: 241
  • DSCF6828.jpg
    DSCF6828.jpg
    109.6 KB · Views: 211
  • DSCF6837.jpg
    DSCF6837.jpg
    102.4 KB · Views: 146
  • DSCF6831.jpg
    DSCF6831.jpg
    92.2 KB · Views: 132
  • DSCF6830.jpg
    DSCF6830.jpg
    111.7 KB · Views: 164
I forgot to add that the rectifier sockets will get replaced, they are pretty lousy at gripping tube pins. Also, I'm thinking about swapping out the rectifier tube from a 5U4 type to a 5V4 type in order to cut some of the current demands on the power transformer and get a nice, slow B+ ramp up.
 
Hi Jim,
It looks to me like the feedback to the input stage cathode is passing through the cathode bypass cap to get to the unbypassed 100 ohm resistor. That puts the 50uF cap in the feedback loop. Seems like connecting the feedback to the top of the 100ohm resistor would be the traditional way.
John
 
Jim -- A couple of things you might consider:

1. I was never aware of a 22"A" version -- the original (and to my knowledge, only) version had pitiful filtering of the B+ supply, using only two paralleled caps at the rectifier tube to filter the output stage B+.

2. A choke would be a wonderful addition to the unit to obtain low noise operation, and could be placed between the two caps mentioned in #1 above. The output stage would connect after the choke.

3. Moving to a 5V4 is also a good move. It will raise the B+ somewhat, but the output stage operates best with 400 volts between the OPT CT and the CATHODES anyway -- which the original design was somewhat shy of supplying. No doubt, the choke will also absorb some of the extra B+ as well.

4. With the improved B+ filtering, anything you can do to raise current flow through the inverter stage will be helpful. The Eico inverter can adequately drive EL34 and 7591 class tubes ok, but is on the ragged edge driving 6L6 class tubes. With the improved filtering, consider cutting the 10K B+ dropping resistor to the phase inverter stage at least in half, and further ensuring that the B+ dropping resistor for the AF amplifier stage then allows for about 95 volts on the plate of the EF86 AF amplifier tube. These measures will increase the drive capability of the inverter stage to ensure it remains well within its limits of linear operation, even under full power conditions.

Finally, conversion of the front end to a Williamson style design can produce even further performance improvements -- completely eliminating the output tube drive issue from the table, and producing superior HF performance over that of the Mullard design.

Good luck with your rebuilds!

Dave
 
Hi John and Dave,

Thanks for your inputs to this thread. Man, it's been a wild week fixing and performing software updates on the Army/Marine combat gunnery simulators I work on. Time to enjoy the weekend and relax!

Both of you zoomed in on the front end of these amps. I think there is a lot that can be gained from reconfiguring the front end with a 6CG7 as a voltage amp/phase splitter and the 6SN7 as a driver setup, ala the Heath W4-AM/W5M. Let me work out some figures and see what I come up with. I'll have to ask the owner if the topology change is ok with him. I think it's a great idea and I bet it would sound better and be quieter as well!

Dave, I have two schematics for the HF22s. One shows the exact configuration you mentioned, two 20uF caps in parallel as the sum of the entire B+ filtering. Rather less than optimum! The other shows a 50 ohm 10 watt resistor between the two 20uF capacitors. This would net a bit improved filtering, but still not very good at all!

I have some 1.5H 38 ohm chokes that will work out pretty well and just fit the inside the chassis, so the original appearance will not be altered at all. I can tweak the value of the first capacitor network feeding the choke to alter the B+ up or down a bit as needed. In any case, I'll use 450V caps in series (with bypass caps and bleeder resistors) on both sides of the choke for some added safety headroom. I just don't like running electrolytic caps within 20 or 30 volts of their max rating. I may go a little overboard, but I do like a trouble-free running amp!

I also forgot to mention that these amps will get screen stopper resistors installed. Dave, that was a great white paper you wrote on the other things that also need to do AFTER you improve a tube amp power supply!

Speaking of the power supply, the other choice for a rectifier tube I was thinking about is the Bendix 6106. Extremely slow warmup (slower than a 6CG3 damper diode!) and the voltage drop across it may be more in line with a 5U4 type. I'll try both and show the results when the rebuilds are done.
 
I had time today to get the first HF22 stripped down, power transformer bells pulled off and transformer inspected, not too bad. No signs of anything bad happening!

The underside of the chassis has some surface corrosion that will be cleaned off, but it's in pretty good shape for it's age!
 

Attachments

  • DSCF6840.jpg
    DSCF6840.jpg
    84.4 KB · Views: 119
  • DSCF6839.jpg
    DSCF6839.jpg
    81.8 KB · Views: 118
Finally, conversion of the front end to a Williamson style design can produce even further performance improvements -- completely eliminating the output tube drive issue from the table, and producing superior HF performance over that of the Mullard design.

Dave

Would this mean that the HF20 (Williamson style) is superior to the HF22 (Mullard style)?

Thanks,

Ron
 
Isolated ground input jack installed. I hate chasing ground loops!
 

Attachments

  • DSCF6841.jpg
    DSCF6841.jpg
    63.4 KB · Views: 117
  • DSCF6842.jpg
    DSCF6842.jpg
    80.5 KB · Views: 103
Here's a shot of the output and power transformers, all finished. The power transformer had some rubber grommets installed on the bells so there will be zero chance of any buzzing. It will also be mounted to the chassis via rubber grommets so no power transformer vibration can be coupled to the chassis or to any tubes.
 

Attachments

  • DSCF6848.jpg
    DSCF6848.jpg
    82.5 KB · Views: 179
  • DSCF6846.jpg
    DSCF6846.jpg
    87.4 KB · Views: 169
Yep! I sure did. The base color is hammertone metallic and the power transformer core stacks are gloss black. 3 coats of color and 3 coats of clear enamel. Each coat baked at 170F for an hour. I have the 9-pin ceramic tube sockets on order.
 
Eico HF-22 Update: Sorry it's been a little bit since I was last able to post an update. Between work and this pesky left shoulder, I haven't been able to do much.

Here's a pic of the underneath of the chassis showing the much needed and now added choke. Also shown is the rectifier socket just about completely wired up and the heater wiring about 50% done. You can also see the ground buss wire. This amp will not use a chassis ground, but rather a ground buss made from pure copper wire. This should have a lower impedance than a steel chassis ground. I hate hum!
 

Attachments

  • DSCF6874AK.jpg
    DSCF6874AK.jpg
    99.6 KB · Views: 171
You have my sympathies about the way work and a pesky shoulder (R in my case) interferes with important stuff like this!
 
You have my sympathies about the way work and a pesky shoulder (R in my case) interferes with important stuff like this!

Brother, thanks and you ain't just a kiddin!!! I've come to terms with the fact that sometime in Nov or Dec I will have to go under the knife and just get this fixed.

Ugh.....
 
Got a little more work done tonight. The heater wiring is complete. The wire is Telfon silver plated twisted pair. Also got the input jack and potentiometer wired in. The second filter section is also shown and consists of two 470uF 450V 105c caps in series take make a 235uF 900V filter. Mounted on the caps are 1M bleeders and .01uF 630V bypass caps. The caps for the first filter section between the choke and rectifier are still on order.
 

Attachments

  • DSCF6876AK.jpg
    DSCF6876AK.jpg
    116.8 KB · Views: 148
  • DSCF6880AK.jpg
    DSCF6880AK.jpg
    78.5 KB · Views: 143
  • DSCF6881AK.jpg
    DSCF6881AK.jpg
    90.2 KB · Views: 125
  • DSCF6882AK.jpg
    DSCF6882AK.jpg
    102.8 KB · Views: 140
Got the third and fourth power supply filter nodes built. I'll wire them in later tonight. Each node consists of a 220uF 450V 105c cap, 1M bleeder resistor and .01uF 630V cap. You can also see the 2 watt 220K dropping resistor that will feed the EF86 plate. Also visible but not wired in is the 10K 10 watt wirewound dropping resistor for the 6SN7 plate supplies.

And, of course, the caps I'm waiting for so I can complete the power supply will no doubt come in on Monday as I am out of town all this coming week on business.
 

Attachments

  • DSCF6884.jpg
    DSCF6884.jpg
    90.6 KB · Views: 121
  • DSCF6883.jpg
    DSCF6883.jpg
    100.3 KB · Views: 114
Back
Top Bottom