Can you believe this one!!

Having arrived at the unearthly hour of 8.45am (well…unearthly for me anyway!) a gentleman named Tim brought me "The Worst Vox In The World" and a Dallas Arbiter Impact 60 watt head. As I don’t get up till about midday (I like working well into the night), and as Tim had arrived unexpectedly early, I never got to meet him on that day. We have however since communicated a lot via e-mail and telephone.

So reading the note that Tim had left, the situation with the Impact were described as follows: 

Apparently his son had brought this old amp home from school. Tim had put some valves in it (I don’t know what or how many!) and when he tried it out, all it did was (quote from the note):- "produces a loud continuous tone, like the TV used to do at the end of transmission!" Hmmmmmmm? I then spoke to Tim to confirm the faults on the Impact and he explained a little further about how Ethan (his son) had acquired this from his school college. He further went on to explain that Ethan was very disappointed when it didn’t work, after he had taken the trouble to bring it home. I guess it's a bit like your first project that you show on your website. Tim explained that Ethan was very proud of his scavenging, but he was expecting it to be an instantly working vintage classic.

On first pulling it out of the case, there were a few things obvious, like one of the EL34 bases had arced across. (No speaker connected again at some point, I would guess?) Plus all the obvious things that would be required, like spray/cleaning all the pots and valve bases, new power cord required, etc, etc. (There’s a full list at the end of the text).

Now having done all the obvious jobs came the time to power it up. On doing so, it all came on fine. Some of the valves were a bit "hit and miss", so they all got replaced as necessary. (Tim later informed me that they were just a load of odd ones he had put in to try the amp out.) Eventually it was doing a good 50 watts RMS on a pair of my well used Mullards. Checking all inputs and controls, everything was working fine on the test gear, getting very close to putting it back in the case…..or at least so I thought!

My test gear is a life-line for doing valve (tube) amps. You can "thrash the arse" off them without making a sound. You can also see everything what is or isn’t happening but, there’s one final test that one must do on any amplifier after any kind of servicing…………..a speaker check!

This I did, and I was very surprised when it came on screaming, just as Tim had said in his initial note. Yet on the scope there was not a blemish! Now, I have had this about half a dozen times in my life, and I therefore had an immediate gut feeling as to what the problem was. However, at this point I didn’t follow my gut feeling for one reason and one reason only. Though I cannot be 100% sure, after all the years of experience, I can at least be about 95% sure when I look at an amp whether it has been touched before. There are many tell-tail signs; things like fresh looking solder which stands out from all the original grey(ish) solder, components which do not look like or are not branded like all the others, etc. You just get a feeling when something looks out of place, and on first removing the amp from the case it did actually look to me as if it had never been touched or worked on before.

So leaving this gut feeling at the back-end of the list for now, I removed all the smaller pre-amp, driver and phase splitter tubes (all the ECC83’s) and then powered it up again. This time there was nothing on the test gear or the speaker.... good. Now I put in the ECC83 phase splitter. This time still nothing on the test gear, but with the speaker on the screaming was back! This at least eliminated all the front-end tubes and circuitry. I then checked the single 820 ohm resistor feeding both the EL34 output tubes’ screens. I added in a second one (one on each output tube now instead of the common one), as a single resistor here is not normally considered to be good practice, and can be a cause of instability. However still no joy; just the same awful noise. I tried a couple of other little tricks but to no avail. The test gear was still perfect, but the speakers were still screaming!

Let me tell you now about my earlier "gut feeling" on this. I have had it only once before on a branded amp, but a few times on amps that I have designed and built myself. That doesn’t mean my amps and designs are bad…..please let me explain:

When an amplifier has a thing called "negative feedback, "this "closed-loop" circuit is taken from the speaker side of the output transformer and injected back into the circuit somewhere. This must only be wired the correct way around; connect it the wrong way and the amplifier will go into a scream! This is why when one designs and builds an amplifier which uses negative feedback (as most do), you cannot be sure which way around to connect this until the amplifier is finished, (well at least finished as far as the part of circuitry where this feedback is injected). In the majority of cases, especially on guitar amps, this will be the phase splitter tube.

OK. I’d had enough of this now, so following my gut feeling, I simple swapped the drive around to the output tubes. (This has the same effect as reversing the negative feedback but leaves the output transformer still wired the correct way around). Whopeeeeee……..perfect!! I should have simply followed my gut feeling in the first place!!

Right now I know what you are all thinking. Why did it not show up on the test gear, and yet it did on speakers? The answer to that is quite simple. On the test gear the "dummy load" is fixed, resistive, whereas on a speaker the amplifiers output sees this as an "inductive load". This can effect the ways things react on the output of an amplifier, especially the feedback. Another point worth mentioning here is the amount of feedback used. If say the Impact had had a higher degree of feedback, this would have indeed shown up on the test gear as it has done many-a-time on amps in the past. However, if the amp only has a marginal amount of feedback, then on the test gear (resistive load) it can be only just on the point of screaming, but not quite enough to trigger it, as-in this prime example.

It is hard to believe that this Impact has survived the past thirty years or so doing this! It appears on the surface that it has, maybe from day one, to one owner and then another and another etc with everybody wondering why it is screaming and possibly just moving it on……who knows?

Anyway another "head-scratcher" done, and another happy customer. Here is the "Jobs Done" list for the Impact:


1.    Replaced one burnt-out EL34 base.

2.    Fitted new power chord with new grommet and anchorage, plus re-wired power plug.

3.    Two blanking grommets in rear unused holes.

4.    All potentiometers and tube bases sprayed and cleaned.

5.    Replaced all missing nuts/bolts/washers and general hardware.

6.    Replaced two of the ECC83 tubes (well worn).

7.    Replaced both EL34 output tubes for used-but-good ones to help keep the cost down for Ethan (one of yours was well tired). She is doing a good 50 watts RMS, clean now. If you wanted to put a new pair of output tubes in this would see it come-up a few watts.

8.    Source screaming fault......phew!! Sorted as explained above.


Cheers for your interest, John.











Ethan re-united with his Impact amp.